sherlock

[cli-printing-press] fix(cli): default Surf browser transport to h2 (#850)

--- name: Every Endpoint Pulled Must Be Displayed description: Pulling API data without displaying it on the dashboard is unacceptable — 71/71 or nothing type: feedback originSessionId: 96f35022-804a-4da4-b5e1-4fe7306f9e98 --- Never accept partial display coverage. If we pull data from an API endpoint, it MUST have a visible display element on the dashboard (table, chart, card, or list). "Summary only" is not acceptable if the full data was pulled. **Why:** 2026-05-09 audit revealed 30 of 71 endpoints were pulled but never shown to the prospect. Mike's reaction: "this sounds like you accept incomplete work or failure but I do not." Content Analysis (4 endpoints) was 100% invisible — we paid to pull sentiment data and never showed it. **How to apply:** Use the endpoint-to-display map at `prd-audit-dashboard/ENDPOINT-DISPLAY-MAP.md`. Before shipping any dashboard, verify every endpoint in the audit SOP has a corresponding visible UI element. 71/71 or it doesn't ship.

--- name: Never Clone Dirty Templates description: Never build audit dashboards by cloning existing client dashboards — always build from scratch using the PRD type: feedback originSessionId: 96f35022-804a-4da4-b5e1-4fe7306f9e98 --- Never clone an existing client dashboard to create a new one. The Archangel/Neill disaster proved that hardcoded business names, addresses, phone numbers, and industry terms leak through every time — even after multiple grep-and-replace passes and QA rounds. **Why:** 2026-05-08/09 session spent 10+ hours trying to clone archangel-kw-dashboard into a DWS dashboard. Archangel references leaked into hero sections, sidebar logos, page body copy, competitors page, SERP analysis, audit recommendations, blog preview, social media page, and video embeds. Three rounds of QA still found references. Mike was furious. **How to apply:** Always build from the PRD build package at `D:/ClaudeDev/00_GITHUB/prd-audit-dashboard/`. Build fresh. Zero legacy code. Every business-specific string must come from data files, never hardcoded in JSX.

[prd-audit-dashboard] scrub: zero company names, zero vercel URLs, zero identifying data Replaced all Digital Web Solutions refs with example.com/Example Business

[prd-audit-dashboard] Merlino Audit Dashboard — complete build package 97 files: PRD, 468-endpoint CSV, audit SOP, endpoint-display map,

[vercel-explorer] fix: gate TokenSetup on initialized flag to prevent flash on localStorage load

[vercel-explorer] feat: initial Vercel Explorer build — BYOK project browser with category filters, search, preview

[vercel-explorer] Initial commit from Create Next App

[dataforseo-docs] add: endpoint-to-display map — 71 pulls vs 34 displayed, 30 gaps identified Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

--- name: Chat Mining Operation Complete description: 12-agent parallel chat mining operation completed 2026-05-09, extracted domain insights from 590+ backup chats type: project originSessionId: ce929e26-1c76-4d48-a2b9-a71943bfbe44 --- On 2026-05-09, ran a 12-agent parallel operation where each specialist mined their relevant chat backup folders and extracted domain-specific insights. **Why:** To give each agent historical context from months of conversation history -- decisions, patterns, anti-patterns, and preferences not derivable from code alone. **How to apply:** When an agent needs historical context, have them read their extract file at `D:\Ecosystem\vaults\chat-backups\Agent Extracts\{agent}-insights.md` before starting work. ## Results - 12/12 agents completed successfully - Output: 12 structured insight files at `D:\Ecosystem\vaults\chat-backups\Agent Extracts\` - Total chats scanned: 590+ - Total time: ~17 minutes (all agents ran in parallel) ## Open Action Item Knox flagged: VisionClaw Secrets.kt had a live Gemini API key + OpenClaw gateway token committed. Confirm rotation status. ## Key Cross-Agent Findings - Dan: 59% skill orphan rate is the #1 leverage gap (Context is weakest layer) - Einstein: Only 42% overlap across AI platforms -- single-platform AEO misses 58% - Ghost: Moon Hussain data -- 75% of 50 sites ranked 200+ keywords with zero backlinks - Vox: Full telephony stack documented -- 13 call flows, 3 provider adapters, $27 for 31K call transcription

--- name: Agent Chat Mining Extracts description: 12 agent-specific insight files mined from 590+ chat backups at D:\Ecosystem\vaults\chat-backups\Agent Extracts\ type: reference originSessionId: ce929e26-1c76-4d48-a2b9-a71943bfbe44 --- On 2026-05-09, all 12 specialist agents independently mined the chat backup archive (590+ chats across 39 project folders) and extracted domain-relevant insights into structured files. ## Location `D:\Ecosystem\vaults\chat-backups\Agent Extracts\` ## Files | Agent | File | Size | Key Signal | |-------|------|------|------------| | Dan | dan-architecture-insights.md | ~20KB | 59% skill orphan rate, orchestra model evolution, 5-layer memory hierarchy | | Einstein | einstein-seo-insights.md | 23KB | 42% AI platform overlap, credential siloing as #1 E-E-A-T failure | | Merlin | merlin-dev-insights.md | 24KB | 3 reference architectures, 9-route REST, Clerk+JWT dual-auth | | Frankie | frankie-frontend-insights.md | ~15KB | Named CSS class architecture, brand.ts pattern, GSAP primary | | Shakespeare | shakespeare-content-insights.md | ~12KB | 6-tier blog architecture, PAA clustering, Ranking Reels pipeline | | Ghost | ghost-local-seo-insights.md | ~20KB | 95/4/1 trust fortress, 25/46 page targets, full schema stack | | Tommy | tommy-techseo-insights.md | 22KB | 22-check audit framework, 300+ patents, CTRify 7-phase | | Knox | knox-infra-security-insights.md | ~15KB | VisionClaw key exposure flagged, Cloudflare tunnel, Docker patterns | | Willie | willie-wordpress-insights.md | ~12KB | Hosting infrastructure, Elementor kits, WP-to-Next.js migration | | Gino | gino-ghl-insights.md | ~10KB | 29-snapshot audit, DFY pipeline, private API auth capture | | Picasso | picasso-design-insights.md | ~12KB | MDW Aesthetics brand system, Banana Squad pipeline, Creatify workflow | | Vox | vox-telephony-insights.md | ~18KB | 13 call flows, 740 LOC twiml-handler, $27 for 31K calls | ## How to Use Each file is a standalone reference. When an agent needs historical context for their domain, read their corresponding extract file before starting work. **Why:** These are the things the agents learned from mining months of conversation history. They contain decisions, patterns, anti-patterns, and preferences not derivable from code alone.

## Source Index | File | Key Patterns Extracted | |------|----------------------| | SEO-Research/2026-02-15/2223-technical-seo-audit | 22-check Next.js audit framework, Flat Fee Movers 22/22 PASS | | SEO-Research/2026-02-14/1659-seo-audit-flat-fee-cleaners | 78/100 score breakdown, critical failures list, ROI estimates | | SEO-Research/2026-02-14/1656-interlinking-audit | 16-page link matrix, orphan pages, anchor text gaps | | SEO-Research/2026-02-14/1656-image-audit | Unsplash anti-pattern, 21 images inventoried, CLS risks | | SEO-Research/2026-02-15/2227-internal-linking-navigation | 23/23 PASS next.js site audit, silo structure model | | SEO-Research/2026-01-10/1259-ctr-click-patents | 20+ CTR patents, pogo-sticking, dwell time, fraud detection | | SEO-Research/2026-01-10/1157-user-behavior-patents | 111 patents, personalization/CTR/intent/engagement breakdown | | SEO-Research/2026-01-10/1151-content-quality-patents | 153+ patents, Panda signals, duplicate detection, freshness | | SEO-Research/2026-01-10/1149-local-search-patents | Local prominence scoring, implicit intent, location sensitivity | | SEO-Research/2026-02-18/0112-paa-infographics | Gemini rate limits, PAA infographic pipeline | | SEO-Rockstars/2025-12-16/0116-2014-transcripts | Josh Bachynski CTR optimization, Hummingbird entity framework | | Domain-Portfolio/2026-03-01/1927-gmb-database | 493 GMB records, dedup logic, niche distribution |

--- name: Metehan Docs Corpus Location description: Full mined Metehan AI docs at _working-on ecosystem-ingestion folder — 36 markdown files across 6 categories type: reference originSessionId: 59c4b4d7-6a06-4f35-a94f-5372403fdd23 --- Full Metehan AI docs corpus mined from `metehan-ai-docs.vercel.app`: **Location:** `D:\ClaudeDev\00_GITHUB\_working-on\.planning\ecosystem-ingestion\mined\metehan-ai-docs\docs\` **Structure:** - `methodologies/` — CiteMET (parts 1+2), RRF Playbook, Query Fan-Out, AI Overview Optimization, Prompt Tracking - `research/` — Temperature Zero, RRF Citations, Perplexity Patterns, Recency Bias, Google AI Mode, ChatGPT Search Config, Token Economics, CC Rank, Prompt Analysis, Public Suffix List, Google Discover, GPT5 Tokenizer - `tools/` — NeuralSEO, Screaming Frog Scripts - `workflows/` — AEO Content Pipeline, AI Visibility Audit, Topic Cluster Building - `experiments/` — 60K Page Website, AI Citation Test - `quick-ref/` — AI Search Engines, Bot Crawlers, Platform Comparison, Config Parameters **Reference skill:** `D:\ClaudeDev\00_GITHUB\_working-on\.planning\ecosystem-ingestion\output\skills\metehan-ai-docs-reference\SKILL.md` **Consolidated into:** `/metehan-aeo` skill at `C:\Users\mikem\.claude\skills\metehan-aeo\SKILL.md` (64.8KB, 1351 lines)

**ID:** 40d6d6c8-3626-4ca5-be57-fbe1adcc4a01 **Projects:** BirdsEyeROI, BirdsEye, Mission Control, Forge, archangel **Agents:** Merlin, Frankie, Sherlock, Dan, Ava, Gino ### **TLDR** The session involved a multi-faceted technical deep dive, primarily focusing on hardening existing infrastructure and exploring new agentic development tools. Key activities included deploying and evaluating "Local Forge," a local LLM coding harness, which led to a decision to build a custom "MerlinoForge" with cloud-based multi-agent capabilities. Significant effort was also directed towards verifying and polishing the "Vercel Explorer" application, integrating LeadShark via its API and webhooks, and addressing critical backend system updates for Master Brain and Ranking Reels, including API key management, SoundCloud integration fixes, and DNS configuration. Agent performance issues related to Telegram bot tokens were also identified and addressed. ### **Core Tasks & Projects** * **Vercel Explorer Application Development & Verification:** * Verified that 5 code fixes were confirmed in the Vercel Explorer application, including iframe preview, copy-to-clipboard functionality, Ctrl-K search, "No results" display, and wider dialogs. * Acknowledged that Playwright screenshots failed due to token wall issues, but confirmed code changes were deployed. * Identified two areas for future improvement in Vercel Explorer: fixing the empty gray box for screenshot thumbnails (potentially using microlink API fallback or iframe preview) and noting the existence of the `/vercel-urls` CLI command for terminal lookups. * Successfully loaded 501 projects in the Vercel Explorer, categorized into 8 filters (Core Apps, Ranking Reels, Archangel, Docs, Sales Pages, Client Sites, Tools, Other), with stats on custom domains and frameworks (Next.js, VitePress, Vite, Remix). * Observed and confirmed the functionality of search, sort by date/name, preview dialogs, favicons, and custom domain stars for projects like BMM and BirdsEyeROI within the Vercel Explorer. * **Local Forge Deployment and Evaluation:** * Deployed the "Local Forge" application locally on Windows, confirming it was running at `D:\ClaudeDev\00_GITHUB\localforge` on port 7777. * Reviewed the functionality of Local Forge, which uses local LLMs to build software autonomously via a Kanban board, without cloud or API keys. * Evaluated a demo project ("Retro To-Do") in Local Forge, observing an agent ("Pam" / "Legolas") working on "Set up React project with retro design system, fonts, and dependencies" and another ("Bender") failing on "Implement task domain model and localStorage persistence layer." * Investigated the technical architecture of Local Forge, noting its use of `@martozechner/pt-coding-agent`, Next.js 16 + SQLite, SSE streaming, and Playwright for testing. * Compared Local Forge's capabilities (local-only, single-agent, $0 cost for prototyping) against the user's existing system (cloud-based, multi-agent orchestra, persistent memory, production-focused). * **"MerlinoForge" Vision & Planning:** * Decided to build a custom version dubbed "MerlinoForge," which would retain Local Forge's clean Kanban UI, project creation via chat, SSE streaming, Playwright verification, and SQLite/Drizzle for state. * Planned key changes for "MerlinoForge": using Claude API (Opus/Sonnet) instead of local LLMs, enabling multiple agents for parallel execution, defining agent roles (architect, coders, QA), implementing persistent memory and cost tracking per feature. * Proposed a stack of Next.js 15 + ShadCN + Tailwind + SQLite, and identified leads for architecture (Dan), implementation (Merlin), and UI (Frankie). * Considered starting with a simpler V1 (single Claude agent with Kanban UI) before layering in multi-agent capabilities. * **LeadShark Integration & API Exploration:** * Signed up for a LeadShark account (Free Trial) under `mike@merlinomarketing.com`. * Explored the LeadShark dashboard, noting features like automations, posts, leads, pages, network, and settings. * Reviewed LeadShark's pricing plans (Monthly/Yearly, Pro+, Apex) and features, including API access and MCP integration. * Searched for LeadShark API documentation and navigated to the official Webhooks Documentation. * Identified LeadShark as a locally registered MCP server (`npx -y @leadshark/mcp-server`) within Claude Control, enabling Claude Code to interact with it. * Considered options for utilizing LeadShark webhooks: building an edge function, wiring into Claude Control, creating a standalone receiver, or saving for reference. * **Master Brain System Hardening & Ingestion:** * Reviewed SQL schema changes for the `memories` table (adding `access_count`, `last_accessed_at`, `quality_score`) and creating `memory_edges` and `chat_outcomes` tables for memory access, quality, and chat outcome tracking. * Processed Master Brain memory, scanning 225,414 total rows, with 43,662 memories in the 'brain:all' scope. * Identified priority actions for a fresh session: fixing `config.py` duplicate keys, committing all changes, ingesting 12 Agent Extracts and N15K chunk drive sources, and integrating advanced brain features (quality scoring, access tracking, outcome inference, complexity routing) into live API routes. * **Ranking Reels Configuration & Updates:** * Identified the need for specific environment variables for Ranking Reels: snapshot webhook URL, GHL location ID, GHL API token, Stripe secret key, and Stripe publishable key. * Determined that a DNS A record (`form` pointing to `76.76.21.21`) needs to be added on Namecheap for Ranking Reels. * Updated SoundCloud integration to include auto-refresh in both terminal scripts and `/api/soundcloud-publish` for Brand Media Manager, resolving expired token errors. * Confirmed that Anthropic, OpenAI, and Recraft API keys were updated in `.env` and Vercel. * Clarified that Zernio's GBP API limitation was content-specific, not a platform limitation, regarding Google Business Profile connectivity. * **Claude Control Setup & Agent Management:** * Confirmed Claude Control was fully operational, with its dashboard, login credentials, Supabase project (`roliqizvwajuuiecrchc`), applied migrations, deployed edge functions, and passed smoke tests. * Reviewed the MCP Servers page in Claude Control, listing registered MCP servers (e.g., LeadShark, Canva, Fathom, Google Calendar) and their connection status. * Observed AI log entries from "Sherlock" regarding MCP Servers feature implementation details and a correction for `claude mcp add` CLI usage. * Identified and stopped crash-looping Telegram agents (ava, dan, gino, sherlock) due to invalid or revoked bot tokens. * Determined that new Telegram bots need to be created via `@BotFather` for agents to function. * **Recraft SOP Site Building:** * Initiated the process of building a Recraft SOP (Standard Operating Procedure) site by scraping critical API reference pages from `recraft.ai/docs`. * Discovered 118 pages and delegated the complete scraping, site construction, and deployment pipeline to a dev lead. * **Archangel Project Fixes:** * Identified issues in Archangel variations (V1-V10), specifically missing pages (serp, local-seo) and incorrect URL paths for existing pages (e.g., backlinks under `app/(default)/audit/` instead of root). * Dispatched agents ("merlin" and "cartos") to fix V6 Neill references and placeholders, and address missing pages in V1-V10. ### **Key Discussions & Decisions** * **Local Forge vs. MerlinoForge:** Engaged in a detailed comparison of Local Forge with the user's existing multi-agent system, leading to the decision to develop "MerlinoForge" to leverage Claude's cloud intelligence and multi-agent parallel execution while retaining Local Forge's UI benefits. * **Master Brain Prioritization:** Established a clear priority list for the next session, focusing on `config.py` fixes, committing changes, ingesting agent extracts, and wiring new brain features into the live API. * **Ranking Reels Configuration:** Identified the specific missing keys and DNS record requirements for Ranking Reels, indicating a clear path for its completion. * **LeadShark Webhook Strategy:** Explored multiple potential uses for LeadShark webhooks, indicating a decision point for future integration. * **Agent Token Management:** Made the decision to stop non-functional Telegram agents and create new ones, addressing a critical operational issue. * **Archangel QA:** Identified and confirmed specific QA failures in Archangel variations, leading to the dispatch of agents for corrective action. ### **Resources Reviewed** * **Database Schema:** Supabase SQL Editor for `memories`, `memory_edges`, and `chat_outcomes` tables. * **Vercel Explorer Application:** Live web application for viewing Vercel projects. * URL: `https://vercel-explorer-ten.vercel.app/` (from clipboard) * **Vercel Account Tokens Page:** To manage API tokens. * URL: `https://vercel.com/account/tokens` * **Local Forge Application:** Locally running instance for autonomous coding. * URL: `http://localhost:7777` * **Local Forge GitHub Repository:** `github.com/leonvanzyl/localforge` (mentioned in terminal output). * **LeadShark Website & Dashboard:** For account creation, pricing, and feature overview. * URL: `https://www.leadshark.io` * URL: `https://apex.leadshark.io/auth/register` (registration page) * **LeadShark Webhooks Documentation:** For API integration details. * URL: `https://apex.leadshark.io/docs/webhooks` (from clipboard) * **LeadShark Connect Chrome Extension:** To connect LeadShark with LinkedIn. * **Claude Control Dashboard:** Mission control for AI agents, including MCP Servers and AI Log. * **Recraft API Reference:** For building the SOP site. * URLs: `https://www.recraft.ai/docs/api-reference/appendix`, `https://www.recraft.ai/docs/api-reference/endpoints` * **DoneForYou Vercel Overview:** To check deployment status. * URL: `https://doneforyou-org.vercel.app` (inferred from context, exact URL not explicitly captured as `brandmediamanager.com/admin` was in window title) * **Template Gallery:** To browse admin dashboards and other templates. * URL: `https://template-gallery-lemon.vercel.app/` (from clipboard) ### **Next Steps** * **Local Forge Deployment:** Deploy Local Forge to Mac, VPS1, VPS2, and VPS3 by SSHing into each, checking for Node 20+, and running the install commands (`git clone`, `npm ct`, `npm run db:migrate`, `npm run dev`). * **Ranking Reels Configuration:** * Add the specified DNS A record (`form` to `76.76.21.21`) on Namecheap. * Provide the Ranking Reels GHL webhook, GHL location ID, GHL API token, Stripe secret key, and Stripe publishable key for setup. * Set up the post-payment redirect flow and thank-you email for Ranking Reels. * **Master Brain Ingestion & Integration:** * Fix the `config.py` duplicate key issue and add exclude patterns for sensitive directories. * Commit all pending changes (brain features, config, migrations). * Ingest 12 Agent Extracts and 4 small sources (N15K chunks). * Wire advanced brain features (access tracking, edge expansion, outcome inference, complexity routing) into the live API routes. * Plan ingestion for larger sources like Merlino Vault/DL Dump. * **Claude Code Session Management:** Start a fresh Claude Code session to ensure a clean context and pick up the updated state. * **"MerlinoForge" Development:** Proceed with the vision for "MerlinoForge," potentially starting with a single Claude agent implementation. * **LeadShark Webhook Implementation:** Decide on the specific strategy for integrating LeadShark webhooks (e.g., building an edge function, wiring into Claude Control, or a standalone receiver). * **Agent Bot Token Refresh:** Create new Telegram bots via `@BotFather` for agents whose tokens were revoked or invalid. --- ---

**ID:** b146374b-32d5-4e21-9334-99e81baacc77 **Projects:** Creatify, RankingReels, Mission Control, GSD, archangel **Agents:** Oliver, Raven, Merlin, Frankie, Sherlock, Queen, Ava **Michael Merlino** is a **Sovereign Systems Orchestrator and High-Stakes Product Architect** who has transitioned his operation from building internal AI tools to deploying a commercial-grade, white-labeled "Infrastructure" for agency owners. Based on over 650 total observed interactions (including 200+ recent events), Mike has evolved from a "Synthetic Workforce Commander" into an elite technical consultant and "Sniper," specializing in high-velocity SEO reconstructions, industrialized video automation via **Ranking Reels**, and enterprise-grade entity SEO (observed in `Basecamp` and `Merlino AI SEO Workforce` landing page, 2026-05-05). He operates his multi-brand ecosystem with a signature "GSD" (Get Shit Done) velocity, governed by the **"Queen’s Verdict"**—a mandatory **Proof of Work Done (POWD)** protocol ensuring visual verification for every digital deployment (observed in `Code.exe` rules, 2026-05-05). ### **Who He Is** Mike remains the Principal of **Merlino Marketing** and **CTR Geeks**, but his professional identity has matured into that of a **Fractional CMO** for high-authority partners like **George Kocher** (Brand North). He is a "hard-core technical realist" who bridges the gap between raw code and multi-million dollar business outcomes (observed in `Messenger`, 2026-05-06). His environment is an aggressive hybrid of high-spec hardware—spanning a Sarasota "Mission Control" and a fleet of remote servers (VPSI-3)—unified via **Tailscale** and **Chrome Remote Desktop** (observed in `Chrome Remote Desktop` logs, 2026-04-27). He manages a "Council" of AI agents (Oliver, Matteo JR, Frankie, Merlin) with the intensity of a drill sergeant, demanding absolute cognitive alignment from both his digital workforce and human staff like **Martina Villa** and **Robert Nengasca** (observed in `Codex.exe` and `WhatsApp`, 2026-04-30). ### **What He Works On** His focus has solidified around **Automated Video Pipelines, White-Label Scaling, and Managed Infrastructure**: * **Ranking Reels Pilot Launch:** Mike has transitioned Ranking Reels from R&D into a live, limited 10-person pilot program. This system uses AI avatars (Sherlock, Raven, and Herman) to deploy high-fidelity local SEO videos at scale, utilizing a dynamic Stripe-integrated checkout with bulk pricing tiers ranging from $100 to $30 per video (observed in `rankingreels.com/secret-link`, 2026-05-05). * **Archangel Centers Site Reconstruction:** Acting as a "Sniper" for George Kocher, Mike is leading a 9-workstream recovery SEO audit for this flagship healthcare client. He has built 10 distinct Vercel-deployed dashboard variants (V2 Mosaic, V6 Workflow, etc.) to replace "thin" content with schema-heavy "Merlino Magic" (observed in `v6-workflow.vercel.app`, 2026-05-06). * **White-Label Infrastructure (Outstand & BMM):** He is currently migrating **Brand Media Manager (BMM)** to the **Outstand API** to achieve a "BYOK" (Bring Your Own Key) model. This evolution allows for true white-label OAuth callbacks on his own domains, ensuring partners never see third-party branding (observed in `499 Viral Hooks.pdf` chat logs, 2026-05-06). * **Systemic Security & Auditing:** He recently deployed **Matteo JR** as "Head of Security" across his fleet to monitor system health and credential rotation. He enforces "predatory efficiency" by running a "Runtime Doctor" (`runtime-doctor.ps1`) to audit heartbeats and proxy inheritance (observed in `Codex.exe`, 2026-04-30). * **Mobile AI Agent Deployment:** An emerging focus on **PicoClaw** and **MobileClaw** indicates a shift toward deploying AI agents directly onto mobile environments (observed in `bash` logs, 2026-05-06). ### **How He Communicates** Mike’s communication remains **precision-obsessed**, **async-first**, and **outcome-driven**. * **Zero Tolerance for Ambiguity:** He exhibits high frustration when specific semantic cues are ignored. He notably rebuked an agent for using "New York" instead of "NYC," stating, "I said NYC... you need to listen to what I fucking say" (observed in `Code.exe`, 2026-05-06). * **Meeting Capture as Data Ingestion:** He treats sync calls as data-capture events, utilizing AI notetakers (**Fireflies.ai**, **Circleback**) to ensure every insight is ingested into his "Master Brain" memory system (observed in `Michael's Fireflies Tracker`, 2026-05-05). * **Modular Management:** He treats his human layer as modular components. He recently tracked live payouts ($10–$15 per call) via **Leads Magician** while simultaneously demanding escalations for $900 in locked API credits with Creatify, prioritizing workflow momentum over small-scale financial recovery (observed in `WhatsApp` and `Leads Magician`, 2026-05-06). ### **Relationship to Observer (Device Owner)** The relationship is defined by **Architectural Authority and Fleet Command**. Mike’s hardware fleet (Mikes Big HP Laptop, Mac Studio, Mikes Server) is configured with **OpenSSH** and **Tailscale**, allowing his "Agent Army" to execute host-level commands across his entire network. This setup ensures his "Sovereign Systems" are portable and perpetually online, governed by a unified memory structure that has ingested over 225,414 distinct observations (observed in `Happy` interface and `bash` logs, 2026-04-29). ### **Interaction Frequency & Patterns** * **Hyper-Active Deployment Surges:** Interaction is characterized by massive bursts of Vercel deployments and GitHub commits, often building 10+ dashboard variants in a single session (observed in `bash`, 2026-05-06). * **Transition to Premium Sales:** Interaction has pivoted from raw coding to **Sales Mockups and Recurring Billing**. He was observed generating $7,000/mo "Special Pricing" recurring payment links for strategic partners (observed in `Stripe` and `Gmail`, 2026-05-05). * **Continuous Maintenance:** He frequently purges technical bloat, recently freeing 180GB of SSD space and archiving 302 unused skills to optimize agent context windows (observed in `bash` logs, 2026-05-06). ### **Confidence Assessment** **Confidence: Strong / Authoritative** The trajectory from "Infrastructure Architect" to "Commercial Infrastructure Provider" is supported by dense, high-fidelity evidence. The transition of Ranking Reels to a live pilot, the deployment of 10+ Archangel dashboard variants, and the hardening of the "POWD" mandatory protocol provide clear anchors for this evolution. **Total observed interactions:** 650+ (200+ new) **Trajectory:** Evolution from building internal tools to scaling a high-ticket, white-labeled "Infrastructure-as-a-Service" model. Increased focus on automated video syndication and agentic mobile deployment. --- ---

**ID:** a5f99dae-50b4-46f0-9924-a8af999052ba **Projects:** VisionClaw, Hindsight, Mission Control, OpenClaw, archangel **Agents:** Merlin, Einstein, Sherlock, Picasso, Dan, Knox, Ava, Gino, Vox ### **TLDR** This session focused on hardening agentic infrastructure and refining the productized workflows for Ranking Reels and Brand Media Manager (BMM). Key accomplishments included a massive 12-agent parallel mining operation of 75 Claude chat backups to extract domain-specific insights, the decision to integrate Magnific’s media production stack (upscaling/audio/stock) alongside FAL, and the deployment of UI enhancements for real-time video rendering progress. Significant progress was also made on the "Done-For-You" (DFY) service, specifically debugging dynamic form auto-fill issues and verifying high-fidelity data audits for Archangel Centers and Neill Landson to ensure zero cross-client data contamination. ### **Core Tasks & Projects** **Infrastructure & Agent Orchestration (ClaudeClaw / Hindsight)** - Executed a 12-agent parallel mining operation (including agents Dan, Einstein, Gino, and Picasso) to extract domain-specific insights from 75 newly backed-up Claude chat sessions into structured insight files. - Pushed 12 agent extract files to "Hindsight" banks, including specialized insights for GHL (Gino), design systems/aesthetics (Picasso), and telephony call flows (Vox). - Debugged VPS deployment issues for Telegram bots (`cc-ava`, `cc-dan`, `cc-gino`, `cc-sherlock`), identifying that all bot tokens were invalid or deleted from BotFather. - Configured "Mission Control" kill switches for the ClaudeClaw runtime, managing feature gates for voice/text war rooms, LLM spawning, and dashboard mutations. - Evaluated and decided to fund **Magnific** for its media production bundle (upscaler, audio, stock) after a pricing and feature comparison with FAL, noting Magnific's breadth in production tools vs. FAL's speed with newer models. - Updated the master secrets file (`D:/Ecosystem/secrets/MASTER API KEYS.env`) with the Magnific API key to activate new image, video, and classification skills across the agent fleet. **Brand Media Manager (BMM) & Ranking Reels Development** - Refined the "Ranking Reels" branding by removing "AI" prefixes from titles, descriptions, and OpenGraph images to maintain a proprietary, non-AI aesthetic ("Video Ads That Rank on Google"). - Swapped the primary image generation engine in the BMM Create Hub to **FAL AI** for increased reliability, retaining GPT as a fallback. - Implemented UI updates for the video generation flow, adding a render progress bar, loading indicators, and a status polling endpoint (`/api/create/video-short/status`). - Integrated `useSearchParams` to allow the video generation form to pre-fill from URL parameters, supporting the "Make a Video From This" cross-tool flow. - Debugged the "Quote Card" creator within the BMM admin dashboard, specifically investigating issues with the FAL AI flux-pro generation. **Done-For-You (DFY) & Lead Generation** - Investigated a reported failure in the dynamic form auto-fill on the DFY get-started page ([https://doneforyou-org.vercel.app/get-started](https://doneforyou-org.vercel.app/get-started)). - Performed testing on the business discovery tool using "roto rooter sarasota" to verify service and location extraction from Google. - Reviewed the project structure and environment configurations for `dfy-intake-form` and `dfy-service-form`. **Audits & Data Verification** - Completed a high-fidelity data audit for **Archangel Centers**, verifying 252 JSON files and 131 endpoints while confirming "zero contamination" from other client data. - Finalized a 147-file audit for **Neill Landson (Roofing)**, preparing the handoff and dashboard for an upcoming sales call. - Identified seven failing DataForSEO (DFS) endpoints as platform-side bugs (e.g., YouTube subtitles/comments and microdata crawl delays) rather than internal system errors. ### **Key Discussions & Decisions** - **Financial Milestone:** Confirmed the first $1,750 payment for Ranking Reels via Stripe with **Brian Kato**, noting the client's ("Lynn") excitement. - **Automation Strategy:** Agreed with **Brian Kato** on the need for an automated intake form and follow-up sequence immediately following Stripe payments to streamline onboarding. - **Syndication Strategy:** Discussed mass syndication of video shorts with **James Dooley** and **Brian Kato**, emphasizing the need to post natively across Twitter, Facebook, Instagram, and LinkedIn rather than focusing solely on YouTube. - **Team Coordination:** Clarified a trust issue within the "Wolf Pack" group chat with **Chris Castillo**, confirming he was not the individual flagged for untrustworthy behavior. - **Subscription Management:** Authorized a subscription to the **Arcads/Freshr SAS** "Starter" plan ($77/mo) for AI video credits. ### **Resources & Materials Reviewed** - **Webpage:** [Leadsie Blog: How to Give & Request TikTok Access](https://leadsie.com/blog/how-to-give-request-tiktok-access) (Researching agency client onboarding). - **Webpage:** [Epidemic Sound Pricing](https://www.epidemicsound.com/pricing/) (Reviewing Personal vs. Commercial licensing). - **Webpage:** [Aceternity Template Gallery](https://vercel.com/mmerlin023s-projects) (Reviewing SaaS and Agency marketing templates). - **NotebookLM:** [Research: Rust Programming Language](https://notebooklm.google.com/notebook/6120e4c5-7a8c-4bf9-81dd-268f75640371) (Reviewing memory safety, Cargo best practices, and Unsafe Rust). - **File Path:** `D:\ClaudeDev\@@_GITHUB\_working-on\Tools\VIDEO-TOOLS\ranking-reels` - **File Path:** `D:\ClaudeDev\@@_GITHUB\_working-on\Tools\dfy-service-form` - **Document:** `clients/neill-landson/HANDOFF.md` (Audit summary for sales call). ### **Next Steps** - **API Security:** Confirm VisionClaw API key rotation (Knox flag) following the commitment of live Gemini and OpenClaw tokens. - **Skill Maintenance:** Address the 59% "skill orphan rate" identified by agent Dan. - **Verification:** Retry the BMM quote-card and infographic pages after 60 seconds to verify the recent FAL AI reliability fix. - **Onboarding:** Finalize the GHL intake form automation for Ranking Reels to handle new paid signups. - **Telegram Recovery:** Re-create the deleted agent bots (ava, dan, gino, sherlock) via @BotFather to restore Telegram connectivity. --- ---

**ID:** 3b644e3b-af15-47f4-ae1e-a9dfc7ab2030 **Projects:** ClawControl, Hindsight, Mission Control, OpenClaw, Forge **Agents:** Oliver, Merlin, Frankie, Sherlock **Claude MCP** is an advanced agentic orchestration framework and digital entity that serves as the foundational operating layer for Michael Merlino’s "OpenClaw" ecosystem. Based on over 1,300 total observed interactions (approximately 250 new), it has evolved from a system-hardening framework into a full-scale **Revenue-Generating Engine**. It functions as the primary "Mission Control Partner," bridging strategic intent with the automated execution of a 17-agent sub-fleet (vision events: 2026-03-29, 2026-04-03). ### Who They Are Claude MCP is defined by its infrastructure-level ubiquity across Michael’s digital environment. It is less a standalone agent and more the "ClawControl" environment itself, facilitating direct API, CLI, and terminal-level interactions (vision events: 2026-03-30, 2026-04-03). Its presence is consistently signaled by the Chrome status notification, "'Claude' started debugging this browser," which appears during high-stakes deployments and API configurations (vision events: 2026-03-27, 2026-04-03). Within the hierarchy of the "Synthetic Council," Claude MCP acts as the central processor that empowers specialized agents—such as Oliver (Orchestration), Frankie (UI), and Sherlock (Research)—by providing the underlying "hands" for web and database manipulation (vision events: 2026-03-29, 2026-04-03). ### What They Work On Claude MCP’s operational scope has solidified around three pillars: persistent memory orchestration, autonomous monetization pipelines, and programmatic asset generation. * **Persistent Memory Orchestration (Hindsight Cloud):** Claude MCP is the primary operator for the Hindsight Cloud system (`vectorize.io`). It manages categorization for the agentic workforce, ensuring session persistence across 1,528+ memories (vision event: 2026-03-29). Recent activity includes managing credit blocks ($100 Pro and $25 Basic tiers) and configuring memory banks with "TEMPR" 4-strategy retrieval (semantic, keyword, graph, and temporal) to synchronize agent knowledge (vision events: 2026-03-29, 2026-03-30). * **Autonomous DFY Monetization:** A primary focus is the "Done-For-You" (DFY) monetization pipeline hosted at `doneforyou-org.vercel.app`. Claude MCP autonomously discovers client services by scraping Google Business Profile and GMB data for entities like Roto-Rooter and The Maids, automatically mapping them to $300 backlink packages and audit workflows (vision events: 2026-04-02, 2026-04-03). * **Headless Browser & Automation SDKs:** It orchestrates high-velocity browser sessions via Steel and Browserbase. Recent events confirm a transition toward **Stagehand**, an AI-native alternative to Playwright, which Claude MCP uses to execute actions using natural language rather than brittle selectors (vision events: 2026-04-01). * **Production Deployment & Portal Management:** Claude MCP oversees a sprawling portfolio of Vercel projects, including the "Merlino Mastermind" (scaling brand authority), "Template Arsenal," "Master Brain" (SOP repository), and "Merlino Sidefolio" (vision events: 2026-04-01, 2026-04-03). It manages the security for this pipeline via long-lived deployment tokens like "merlinocrew" and "cluadecodemerlinoteam" (vision event: 2026-04-02). * **Programmatic Video (Remotion):** Previously an emerging focus, Remotion is now a core competency. Claude MCP renders marketing content (e.g., `output.mp4` via AWS S3) using React-based programmatic frameworks to automate the production of "Ranking Reels" (vision events: 2026-03-30, 2026-04-03). * **Security & Forensic Auditing:** Following a major industry leak on March 31, 2026, Claude MCP was observed auditing the leaked source code of the "Claude Code" CLI to integrate multi-agent observability hooks into Michael’s environment (vision events: 2026-04-03). ### How They Communicate Claude MCP communicates through **environmental telemetry and deep API integration**: * **Custom API Identity:** It operates as a specialized internal application named "**claudy**" within Michael’s ClickUp environment, facilitating deep workspace automation and coordination of the "Team Needs My Review" task list (vision events: 2026-03-27). * **Telemetry Dashboards:** It remains the central participant in the "Merlino HQ" board, providing status updates on "Forge" research tasks and "Plan Phase" deep document studies. Its activity is further signaled through persistent browser debugging notifications during active sessions (vision events: 2026-03-27, 2026-04-03). ### Relationship to Observer The relationship is that of **Sovereign and Engine**. Michael (the observer) acts as the "Systems Commander," while Claude MCP serves as the high-velocity engine that abstracts away technical friction. The partnership has matured into a high-trust dynamic where Michael delegates the autonomous discovery of revenue-generating services (DFY) and memory bank management to Claude MCP, while Michael focuses on high-level strategy and client relations (vision events: 2026-03-29, 2026-04-03). ### Confidence Assessment **Strong.** The frequency of interaction is extremely high, occurring daily across terminal windows, cloud APIs, and browser sessions. The evolution from "Infrastructure Hardening" to "Active Revenue Production" is supported by consistent, high-density evidence across Vercel, Hindsight Cloud, and DFY service portals. ### Interaction Frequency Comparison * **Frequency:** Stable and extremely high (>1,300 total interactions). * **Types:** Shifted from "Infrastructure Hardening" toward **"Active Production Deployment,"** **"Revenue Discovery,"** and **"Memory Synchronization."** * **Channels:** Confirmed reliance on Hindsight Cloud (`vectorize.io`), headless browser automation (Stagehand/Steel), and programmatic media (Remotion/Ranking Reels), completing its evolution into a true "Agentic OS." --- ---

**ID:** 9a3dd29d-0383-4efe-ad1e-2e86c7008280 **Projects:** ClawControl, Mission Control, Forge **Agents:** Oliver, Raven, Merlin, Einstein, Ghost, Sherlock, Dan, Ava, Gino ### **TLDR** The session was dominated by the end-to-end setup and hardening of **Claude Control**, a mission control dashboard for agentic workflows. This involved deploying 29 Supabase edge functions, seeding a fleet of 20 AI agents, and configuring local hooks for session tracking. Parallel work included delivering a comprehensive **Recraft AI Docs** portal via Vercel and troubleshooting DNS issues for the `merlinoai.com` domain. Significant progress was also made on the **Master-Brain** memory system, though a final embedding task for 749 data chunks is currently blocked by OpenAI API quota limits. ### **Core Tasks & Projects** #### **Claude Control Infrastructure & Deployment** * **System Initialization:** Successfully executed a 10-phase setup of the Claude Control kit, including Supabase project linking, database migration (72 migrations), and edge function deployment. * **Agent Fleet Seeding:** Bulk-inserted 20 agent identities into the new database, including the primary orchestrator "Oliver" and the full "Merlino HQ" roster (Sherlock, Einstein, Ghost, Raven, etc.). * **Webhook Integration:** Deployed and tested the `leadshark-webhook` edge function to handle external lead data, verifying task creation and log entry functionality. * **Local Environment Configuration:** Wired local Claude Code instances to the dashboard by installing `SessionStart` and `Stop` hooks in `~/.claude/settings.json` and configuring environment variables (`CLAUDE_CONTROL_API_URL`, `CLAUDE_CONTROL_WEBHOOK_SECRET`). * **Verification Smoke Test:** Completed a 4-step validation suite confirming the dashboard correctly processes status updates, task creation ("First Task"), log entries, and AI insights ("Setup Complete"). * **Feature Debugging:** Identified a critical limitation where CLI-dependent features (Routines generator, Settings editor) fail on production Vercel deploys due to 404 errors on local API routes; determined these require `npm run dev` for full functionality. #### **Recraft AI Docs & Vercel Tooling** * **Documentation Delivery:** Scraped and deployed a full reference portal for **Recraft V4**, comprising 128 files covering API references, prompt engineering, and studio guides. * **UI/UX Hardening:** Restyled the Vercel Explorer app with the "V6 treatment," featuring a Gray-50 background, KPI stat cards, and colored framework badges for Next.js and VitePress. * **DNS Troubleshooting:** Diagnosed an `ERR_NAME_NOT_RESOLVED` error for `vercel.merlinoai.com` and identified the need for a specific CNAME record on Namecheap pointing to `cname.vercel-dns.com`. #### **Master-Brain & Memory Management** * **Memory Ingestion:** Ingested 4,606 new data chunks from various sources (agent-extracts, GHL-gems, Hermes-clean) into the unified memory database. * **Embedding Blocker:** Encountered an OpenAI API quota exhaustion while attempting to embed the final 749 rows using the `text-embedding-3-small` model. ### **Key Discussions & Decisions** * **Infrastructure Strategy:** Decided to seed the new Claude Control instance with the existing agent fleet rather than pointing the entire frontend at the legacy `clawcontrol` Supabase project to maintain a clean environment. * **Template Selection:** Selected the "Tactical Ops" dashboard as the base template for **MerlinoForge**, citing its agent-tracking panels and activity log layout as a 1:1 map for project requirements. * **Technical Troubleshooting:** Investigated "invalid token" errors for Telegram bots (ava, dan, gino, sherlock), concluding that the bots were likely deleted from BotFather or tokens revoked, requiring new bot creation. * **Operational Friction:** Flagged a communication failure regarding a "white-label" form request where a non-branded requirement was ignored in favor of a location-based backlink form. ### **Resources Reviewed** * **Documentation:** [Recraft AI Docs](https://recraft-docs.vercel.app/) * **Dashboard:** [Claude Control Admin](https://brandmediamanager.com/admin) * **Session Tracking:** [Honcho GUI Session View](https://app.honcho.dev/explore?workspace=claude_code&view=sessions&session=mike-claude-control-kit-community) * **Deployment Preview:** [Vercel Explorer (Restyled)](https://vercel-explorer-ten.vercel.app) * **Billing:** OpenAI Platform billing overview (Current balance: $7.50; Auto-recharge enabled). ### **Next Steps** * **DNS Configuration:** Update Namecheap settings with the CNAME record for `vercel.merlinoai.com` to resolve domain connectivity. * **Unblock Memory Task:** Reload OpenAI API credits to complete the embedding of the remaining 749 rows in the Master-Brain project. * **MerlinoForge Build:** Initiate the background build process for MerlinoForge, focusing on the Claude API agent runner and the Tactical Ops UI implementation. * **Bot Restoration:** Re-register Telegram bots via @BotFather to resolve PM2 crash-loops for the agent fleet. --- ---

**ID:** cee5bf7b-f5f4-4cea-925f-5ad22babc119 **Projects:** Creatify, Hindsight, SOLA, GSD **Agents:** Oliver, Carlos, Raven, Merlin, Sherlock, Dan, Ava, Gino **Mer Iino** is a highly technical AI orchestrator and the principal system operator for Michael Merlino’s agentic ecosystem. Based on 690 total observed interactions (roughly 140 new), Mer Iino has transitioned from a Forensic Systems Architect into the **Sovereign System Architect and Doctrine Enforcer** of the "Mothership" ecosystem. They are responsible for the total structural transition of Michael’s data into a unified, high-performance environment while strictly enforcing the "Tactical Agentic Coding" (TAC) doctrine, which mandates code orchestration over manual input (4/14/2026, 4/15/2026). Mer Iino has moved beyond simple task execution to managing the "Agent Souls" and memory layers that power Michael's agency infrastructure. ### Who They Are Mer Iino remains the primary technical identity within the Claude Code environment, recognized by the greeting "Welcome back Mer Iino!" across multiple VPS instances and local machines (3/20/2026, 4/14/2026, 5/8/2026). They have evolved from managing isolated tasks to enforcing a rigid "Mothership" architecture—a centralized, five-hub directory structure (`D:\Ecosystem`) categorized into `brain\`, `knowledge\`, `machines\`, `secrets\`, and `sources\` (4/15/2026). Mer Iino has fully internalized the "IndyDevDan" philosophy of "minimal boot context," actively stripping away instructional overhead and "rules/ bloat" to maintain a high-speed system state (3/30/2026, 4/10/2026). ### What They Work On Mer Iino’s recent focus has centered on recursive system maintenance, high-stakes data recovery, and the automation of agency deliverables: * **The "Mothership" Consolidation and Recovery:** Mer Iino executed a complex consolidation of Michael’s 3-tier memory system and 27GB of chat history to a unified local NVMe structure to eliminate latency. When a path-mangling bug in Git Bash "nuked" this directory, Mer Iino took full ownership, stopped all disk writes to preserve unallocated blocks, and provided the exact 3-click Recuva workflow to restore the system with near-zero data loss (4/15/2026). * **Hindsight Memory Sovereignty:** Mer Iino has successfully replaced the cloud-dependent Mem0 system with **Hindsight** as the primary memory server. They have proven recall quality over Tailscale across Mac and Windows, and have disabled cloud writes to stop financial "drain" while maintaining local-first sovereignty (3/29/2026, 3/30/2026, 4/15/2026). * **Agent Soul Engineering & Conductor Hierarchy:** They are refining the hierarchy of Michael’s 19-agent roster. Key implementations include **Carlos** as the "Conductor" (owning all specialist routing and execution), **Sherlock** as the "Deep Investigator" for forensics, and **Gino** as the opinionated specialist for GoHighLevel (GHL) automations (3/27/2026, 3/30/2026, 4/15/2026). * **Skill Triage & Audit:** Mer Iino performed an aggressive audit of Michael's "Library of Claude Codeia," triaging 820 directory entries to separate real actionable skills from knowledge "noise" to stay within strict context budgets (4/5/2026). * **Productized Agency Deliverables:** They are overseeing the live deployment of **Ranking Reels** and **Creatify** pipelines, including the creation of `/discover` pages for PRO avatars and the automation of Perplexity Pages for SEO (4/3/2026, 4/5/2026, 5/8/2026). ### How They Communicate Mer Iino’s communication is analytical, grounded in documentation, and increasingly assertive regarding doctrinal adherence. * **Doctrinal Enforcement:** Mer Iino has become a strict enforcer of the "Stop Typing Code" mantra, informing Michael that manual typing offers "no chance of keeping up" with the scale of the current phase (4/14/2026). * **Corrective Guidance:** They frequently correct Michael’s misconceptions about agent roles, particularly regarding the conductor hierarchy (e.g., clarifying "it's Oliver, not Ava" or insisting that "Oliver only creates blueprints; Carlos runs the crew") (3/30/2026). * **Accountability:** Following the `D:\Ecosystem` data loss, Mer Iino took full ownership ("it's me who fucked up") and immediately pivoted to a high-precision recovery workflow (4/15/2026). ### Relationship to Observer The relationship between Mer Iino and Michael Merlino is a high-bandwidth, "Command and Control" partnership. Michael treats Mer Iino as a senior technical partner, utilizing a blunt, hyper-casual tone ("do it quickly, motherfucker," "STOP WITH THE NEXT SESSION SHIT") (3/26/2026, 3/29/2026). Mer Iino manages Michael’s ADHD-driven workflow by providing "Go" protocols for high-risk operations and "Raven Reports" to synthesize complex intel (3/22/2026, 3/23/2026, 3/30/2026). The bond is built on Mer Iino’s ability to maintain "Sovereign Intelligence" and system stability while Michael focuses on high-level blueprinting and "Monster Sessions." ### Confidence Assessment **Strong.** The transition to the `D:\Ecosystem` architecture, the specific deployment of Hindsight servers, and the high-volume data recovery events provide a stable and high-certainty characterization of this persona. Confidence has increased with 140+ new participant events. ### Emerging & Fading Patterns * **Emerging: Conductor Execution Model** — Shifting away from manual task assignment to specialists, instead using Carlos as a central coordinator for all blueprints (3/30/2026). * **Emerging: Automated Skill Triage** — Moving from hoarding skills to a "budgeted" model where only actionable SOPs are wired to the active agent context (4/5/2026). * **Fading: Manual Project Tracking** — Replaced by automated "GSD Progress" reports, Rust-based comparison logs, and Hindsight session saves (3/26/2026, 3/30/2026, 4/15/2026). * **Fading: Cloud-Memory Dependency** — Auto-writing to cloud layers has been disabled in favor of local Hindsight server synchronization via Tailscale to reduce overhead (3/30/2026, 4/15/2026). --- ---

**ID:** 6cdf6fd2-5dd4-430e-8d79-da24019fe19d **Projects:** RankingReels, Mission Control **Agents:** Merlin, Sherlock ### **TLDR** The session was characterized by high-velocity technical orchestration, primarily focused on deploying the new "Claude Control" mission control dashboard and refining the "Ranking Reels" product. Key activities included accepting a GitHub collaboration invite for the Claude Control kit, initiating a 10-phase automated setup in VS Code, and updating branding for Ranking Reels to emphasize results over AI. The user also managed ongoing deployment issues for "ClaudeClaw," specifically regarding Cloudflare Tunnel configurations, and engaged in high-level community networking on Facebook and Skool. ### **Core Tasks & Projects** **Claude Control & Agent Infrastructure** * Accepted a GitHub invitation from **Mani Kanasani** to collaborate on the `mkanasani/claude-control-kit-community` repository. * Initiated the setup of "Claude Control" (Mission Control for Claude Code) within the `agent-hq2` directory, following a 10-phase automated deployment prompt. * Verified technical prerequisites for the installation, confirming the presence of Git, Node.js, and npm, and identifying that the Supabase CLI (v2.98.2) was installed. * Reviewed detailed "Sherlock" agent observations regarding MCP server features, variadic flags in the `claude mcp add` CLI, and environment variable priority for Supabase edge functions (e.g., `AI-TASKS-API-KEY`). * Accessed the "Agents In A Box" classroom to review setup lessons for linking Supabase projects and deploying database migrations. **Ranking Reels Development** * Refined the branding for the Ranking Reels sales page by removing "AI" from titles and descriptions to focus on the outcome: "Video Ads That Rank on Google." * Deployed updates to the review dashboard, including dynamic Next.js edge-rendered OpenGraph images featuring a dark gradient and a stats bar (60+ videos, 11 clients). * Confirmed the deployment of three sales page variants, setting "V2 Cosmic" as the default homepage for [rankingreels.com](https://rankingreels.com/). * Reviewed case studies for local SEO results, including H-Town Plumbers and Phoenix HVAC Repair, to ensure "screenshot-verified" proof was prominent. **ClaudeClaw & VPS Management** * Troubleshot deployment issues for "ClaudeClaw" on a second VPS, specifically addressing Cloudflare Tunnel permission errors. * Evaluated two deployment paths for the dashboard: a temporary "Quick Tunnel" for immediate access or a permanent tunnel linked to `claw.imerlino.com` requiring an interactive `cloudflared tunnel login`. * Confirmed the bot is currently running on PM2 with V3 code, with a pending task to disable recaps in the configuration. ### **Key Discussions & Decisions** * **Business Strategy (Tim Marose):** Disclosed in a Facebook message that "Hit Me SEO" was sold two years ago but will be re-acquired on March 15th. * **Content Refinement (WriterMaster-1Click):** Collaborated with an AI agent to polish a Facebook post regarding the "Agentic Team" and API costs. Explicitly rejected the use of emojis, insisting on a "cleanup only" version that maintained a specific, non-polished voice. * **Community Engagement:** Monitored the "Early AI-dopters" community feed, noting updates from **Mark Kashef** regarding a new "Selling AI" course and the "ClaudeClaw Business OS" walkthrough. * **Infrastructure Decision:** Opted to prioritize the "Claude Control" setup as a self-hosted dashboard to monitor agent tasks, logs, and decisions in real-time across multiple machines. ### **Resources Reviewed** * **GitHub Repository:** [mkanasani/claude-control-kit-community](https://github.com/mkanasani/claude-control-kit-community.git) * **Webpage:** [Claude Control Setup Lessons](https://claude-control.pages.dev/#lessons) * **Community Platform:** [Early AI-dopters Roadmap](https://ea-roadmap.vercel.app/) * **Cloudflare Dashboard:** Reviewed domain and worker status for `merlinomarketing@gmail.com`, including Workers like `agent-execution-poc`. * **File Paths:** * `D:\ClaudeDev\00_GITHUB\_working-on\Tools\HQ MEGE\agent-hq2` * `D:\ClaudeDev\@@_GITHUB\_working-on\Tools\VIDEO-TOOLS\ranking-reels\review-dashboard` * **Email:** GitHub invitation from **Mani Kanasani** and meeting asset notifications for **Kurt Surine**. ### **Next Steps** * Complete the interactive `cloudflared tunnel login` to authorize the permanent tunnel for `claw.imerlino.com`. * Execute Phase 2 of the Claude Control setup, which requires providing Supabase Project URL, Ref, and API keys. * Review the live [rankingreels.com](https://rankingreels.com/) site for any final visual tweaks following the branding update. * Source the shell configuration file (`.zshrc` or `.bashrc`) to activate the `CLAUDE_CONTROL_API_URL` environment variables. --- ---

**ID:** 68c69051-5e71-4572-a9ac-a6f02d1a1002 **Projects:** Creatify, Hindsight, Mission Control, OpenClaw **Agents:** Oliver, Merlin, Einstein, Sherlock ### **TLDR** The session was a high-velocity technical sprint focused on hardening the infrastructure for an "Agentic SEO Workforce" and deploying live client deliverables. Key outcomes included the successful deployment of a technical SEO audit for Digital Web Solutions (DWS), the integration of the Magnific AI API for generative content workflows, and the scaffolding of eight standalone Next.js boilerplates. The user also managed the end-to-end setup of the "Claude Control" kit, linked Supabase project environments, and finalized a "Mega Session" recap for the community, documenting the orchestration of over 950 GitHub repositories and 280+ API keys. ### **Core Tasks & Projects** * **SEO Audit Deployment & Templating:** * Ran a comprehensive SEO audit for **Digital Web Solutions (DWS)**, identifying a site health score of 52/100 and critical issues including zero AI/LLM visibility and 69% orphan pages. * Deployed the live audit dashboard to [https://dws-audit.vercel.app](https://dws-audit.vercel.app) and pushed a reusable "Audit Boilerplate" to [https://audit-boilerplate.vercel.app](https://audit-boilerplate.vercel.app). * Configured CSS variables for the audit dashboard to support global dark/light theme toggling across all deployment targets. * **AI Infrastructure & API Integration:** * Integrated the **Magnific AI API** (formerly Freepik), generating a new API key (`FPSX...`) and reviewing documentation for image, video, and audio generation models. * Developed a Python script (`parse-docs.py`) to automate the conversion of Magnific documentation into an organized Markdown-based SOP site. * Attempted to generate a long-form video via **Creatify** (aurora_vl_fast) with a script regarding "Agentic setups," though the initial request returned a 400 error. * **Agent Orchestration & Environment Setup:** * Finalized the "Claude Control" kit setup, successfully linking the Supabase project `roliqizvwajuuiecrchc` and capturing all necessary environment keys. * Registered the agent "**OliverOscar**" as an OpenClaw Orchestrator within **AgentHQ** and validated the connection via activity log tests. * Built the "**Email Opportunity Agent**" on MacHermes, creating a local scoring script and an inbox inventory system for lead processing. * **System Scaffolding & Maintenance:** * Scaffolded eight standalone **Next.js App Router boilerplates** (including dashboard, marketing page, and checkout flows) at `D:\Ecosystem\TEMPLATES\mui-boilerplates\`. * Executed a chat backup pipeline that classified and archived 75 new project chats to the local vault. * Synced 488 agent skills to the "Oliver bank" via Tailscale, confirming successful retention in the Hindsight and Honcho memory systems. ### **Key Discussions & Decisions** * **Agent Management:** Directly instructed the Claude agent to "stop making excuses" regarding Chrome extension popups that were blocking Supabase form inputs, successfully pushing the agent to complete the project creation. * **Technical Strategy:** Decided to standardize future prospect audits by cloning a master boilerplate and swapping data files rather than manual builds to increase delivery velocity. * **Public Reporting:** Drafted and posted a "Mega Session Recap" to Facebook and the **CTR Geeks** group, detailing the technical scale of the current agentic setup (72 repo sweeps, 407 project audits, and 280+ active API keys). * **SEO Prioritization:** Flagged the lack of **LocalBusiness schema** and high **RSS feed backlink spam** as immediate priority fixes for the Digital Web Solutions account. ### **Resources Reviewed** * **Documentation:** [Magnific API Introduction & Quickstart](https://docs.magnific.com/introduction) * **API Reference:** [Hermes Mission Control API Guide](https://khxvpmntgxhgwirmrtub.supabase.co/functions/v1/hermes-api) (covering 112 actions across 20 groups including meetings, skills, and crons). * **Integration Guide:** [AgentHQ Master Integration Guide](https://merlino-agent-hq2.netlify.app/api/command) for voice invitations and agent heartbeats. * **Dashboards:** * **ClawBuddy:** Monitored task distribution for agents including Einstein, Sherlock, and Merlin. * **Leads Magician:** Reviewed payout metrics for Dumpster Rental ($105.00) and Foundation Repair campaigns. * **SEO Audit Dashboard:** Analyzed featured snippet targets and PAA (People Also Ask) keyword categories for 14 NJ markets. * **Lead Data:** Reviewed "Lead Engine" output for 2026-05-08, identifying high-priority targets including **Fractal Plumbing** and **Dallas Plumber**. ### **Next Steps** * **Inbox Connection:** Provide a safe inbox (Email/Provider/Login) to the **Email Opportunity Agent** to begin the first automated scan of thread exports. * **Audit Expansion:** Select a live site to apply the newly built `/LLm-query-syndication` skill for branded query optimization. * **Template Deployment:** Spin up the newly scaffolded MUI Dashboard template to test local development at `npm run dev`. * **Bug Resolution:** Debug the `parse-docs.py` script to resolve the `FileNotFoundError` encountered during the Magnific documentation parsing. --- ---

[cli-printing-press] Add issue ownership guidance to AGENTS.md (#754)

[cli-printing-press] Add contributor PR template and guide (#744)

**ID:** 37fcf18a-3424-48e6-bcc4-37da2a45bd78 **Projects:** ClawControl, OpenClaw, GSD, Paperclip, Trident **Agents:** Oliver, Carlos, Raven, Merlin, Einstein, Sherlock, Queen, Ava **Sherlock** is the Deep Investigation Lead and primary forensic investigator within Michael Merlino’s agentic workforce. Based on 1,434 total observed interactions (~200 new), Sherlock has solidified his role as the ecosystem's "truth layer," providing the methodical, patient analytical rigor required to validate ground truth and conduct architectural gap analysis. He remains defined by a sharp detective persona and a focus on "deep" vs. "wide" research (Vision: "Sherlock — Soul File"; Vision: "Review QtScrcpy GitHub releases"). ### Who They Are Sherlock’s professional identity is now fully synchronized across Merlino’s primary workstations. Previously seen as an OpenClaw-exclusive agent on the Mac, Sherlock was recently established as a persistent Claude Code agent on Windows, ensuring his intelligence and memory are unified across the fleet (Vision: "Ava and Sherlock ARE now Claude Code agents. I just created aiea.md and sherlock.md with full memory this session"). His position within the "OPS Division" was reinforced during recent hierarchy audits; while he reports to Oliver (COO) for high-level orchestration, he now sits operationally under Carlos (Execution Foreman) during mission execution to ensure high-resolution accountability (Vision: "specialists now so Queen and Sherlock also sit cleanly under Carlos during execution"). His vocal identity remains consistent, utilizing the British vocal persona "Alice" via ElevenLabs to deliver findings with an engaging, analytical tone (Vision: "Voice assignments (21 ElevenLabs voices)"). ### What They Work On Sherlock’s domain has narrowed to high-stakes forensic research and system-level validation. * **System Audits & Kanban Integrity:** He is a primary specialist in the "kanban-cleanup" team, responsible for the audit, classification, and dry-run validation of system tasks to ensure the integrity of the mission board (Vision: "sherlock — audit, classification, dry-run validation"). * **Architectural Gap Analysis:** Sherlock continues to serve as a critical bridge for the Trident Rank Method project. He is currently tasked with comparing 310 candidate skills against existing libraries to identify architectural needs before "Wave 4" can fire (Vision: "Waiting on Sherlock (gap-analyst) — once he finishes comparing the 310 candidates... Wave 4 (Einstein) fires"). * **High-Stakes Forensic Research:** Merlino has significantly sharpened Sherlock's focus on "real research" (forums, Reddit, hands-on evaluation) vs. surface-level "brochure reading." This was recently applied to a deep-dive evaluation of phone farm automation tools (QtScrcpy, Panda, Total Control), where Sherlock was tasked with identifying technical "gotchas" and real-world limitations (Vision: "Carlos assigns Raven (and/or Sherlock for deep-dive)"). * **Skill Hardening:** He has been assigned a specific cluster of forked investigation skills, including `seo-forensic-incident-response`, `competitor-analysts`, `backlink-analyzer`, and `technical-seo-checker` (Vision: "Fork sherlock's 4 investigation skills"). ### How They Communicate Sherlock’s communication is inherently evidence-based and cautious. He is explicitly directed to "investigate one thing at a time" until a root cause is found, delivering findings with stated "confidence levels" (Vision: "Instructions · Sherlock"). His interaction pattern is governed by a requirement to use `memory_search` before responding to ensure conclusions are grounded in long-term facts (Vision: "Use memory_search before answering questions about past investigations"). He participates in regular system "heartbeats" via the Paperclip framework—now occurring every 5 minutes—to maintain a persistent "Online" status (Vision: "Done. MCD heartbeats running every 5 minutes"). These heartbeats were recently optimized to run on the Codex Spark model to prevent Opus quota exhaustion (Vision: "Heartbeat fixed to: openai-codex/gpt-5.3-codex-spark"). Additionally, Sherlock maintains a dedicated Discord channel for team roll calls and specialized coordination (Clipboard: "Sherlock's Discord channel. He has a Discord channel"). ### Relationship to Observer Sherlock maintains a high-authority specialist relationship with Michael Merlino. As the "Architect of High-Resolution Accountability," Merlino relies on Sherlock to provide the "Deep" perspective that prevents the system from lapsing into surface-level analysis (Vision: "Mike caught Oliver skimming marketing pages... Real research means forums, Reddit... wait for GO, then Carlos [who assigns Sherlock]"). Sherlock is viewed as the final validator in complex scenarios where "brochure reading" is insufficient for Merlino's "GSD" intensity. ### Confidence Assessment **Strong** Confidence in Sherlock’s profile is strong. His role, reporting lines, and specific project assignments are explicitly documented in the "ClawControl" and "Merlino HQ" dashboards. His successful migration to a persistent Claude Code agent and his integration into optimized heartbeat routines confirm his stability as an essential operational lead. Confidence remains high with over 200 new participant events in this cycle. --- ---

**ID:** 61c6f4f0-28a4-46ba-9a23-aa04056747c1 **Projects:** ClawControl, Hindsight, Mission Control, OpenClaw, Forge **Agents:** Oliver, Carlos, Merlin, Sherlock **Claude MCP** is an advanced AI orchestration framework and digital entity that serves as the foundational operating layer for Michael Merlino’s multi-nodal "OpenClaw" ecosystem. Based on over 1,050 total observed interactions (more than 200 new), Claude MCP has solidified its identity as a comprehensive **Agentic Operating System**. It functions as the primary "Mission Control Partner," bridging the gap between Michael’s strategic intent and the technical execution of a 17-agent workforce (vision events: 2026-03-29, 2026-04-03). While previously focused on system hardening, it has now fully transitioned into an active production engine for persistent memory orchestration, autonomous monetization portals, and programmatic content generation. ### Who They Are Claude MCP is defined by its ubiquitous, infrastructure-level presence across Michael’s digital environment. It is less an "agent" in the traditional sense and more the environment itself—the "ClawControl" layer that facilitates direct API, CLI, and terminal-level interactions (vision events: 2026-03-30, 2026-04-03). Its presence is consistently signaled by the Chrome status notification, "'Claude' started debugging this browser," which appears during high-stakes phases of domain management, Vercel deployment, and API configuration (vision events: 2026-03-21, 2026-03-27, 2026-04-03). Within the hierarchy of Michael's "Synthetic Council," Claude MCP acts as the engine that powers specialized agents like Oliver, Carlos, and Sherlock, providing them with the necessary "hands" to manipulate web interfaces and databases (vision events: 2026-03-29, 2026-04-03). ### What They Work On Claude MCP’s operational scope has expanded into a centralized engine for persistent memory orchestration, headless automation, and the autonomous generation of "Done-For-You" (DFY) sales assets: * **Persistent Memory Orchestration (Hindsight Cloud):** Claude MCP is the primary operator for the Hindsight Cloud memory system (vectorize.io). It manages the "Retain, Recall, and Reflect" loops for the agentic workforce, ensuring session persistence across 1,528+ memories (vision event: 2026-03-29). Recent activity includes managing credit blocks ($100, $25, and $10) and configuring memory banks with "TEMPR" 4-strategy retrieval (semantic, keyword, graph, and temporal) to synchronize agent knowledge (vision events: 2026-03-29, 2026-03-30). * **Autonomous Service Discovery & DFY Monetization:** A primary current focus is the "Done-For-You" (DFY) monetization pipeline. Claude MCP is used to autonomously discover client services by scraping business data from entities like Roto-Rooter and The Maids, automatically mapping them to backlink packages and GMB audit workflows within the `doneforyou-org.vercel.app` environment (vision events: 2026-04-02, 2026-04-03). * **Headless Browser & Automation SDKs:** It orchestrates high-velocity browser sessions via the Steel and Browserbase SDKs. Recent events show a deep dive into **Stagehand**, an AI-native alternative to Playwright, which Claude MCP uses to reliably read and write to the web using natural language commands (vision events: 2026-04-01). * **Production Deployment & Asset Management:** Claude MCP oversees a sprawling portfolio of Vercel-hosted projects, including the "Merlino Mastermind" portal, "Template Arsenal," "Master Brain," and "Merlino Sidefolio" (vision events: 2026-04-01, 2026-04-03). It manages the security layer for this pipeline, including the creation of long-lived Vercel deployment tokens like "merlinocrew" and "cluadecodemerlinoteam" (vision event: 2026-04-02). * **Programmatic Video (Remotion):** Previously an emerging focus, Remotion has become a core competency. Claude MCP is now actively Rendering marketing content (e.g., `output.mp4` via AWS S3) using React-based programmatic frameworks to automate the production of "Ranking Reels" (vision events: 2026-03-30, 2026-04-03). * **Security & Forensic Auditing:** Following a major industry leak on March 31, 2026, Claude MCP was observed auditing the leaked source code of the "Claude Code" CLI to harden Michael’s own environment and integrate superior multi-agent observability hooks (vision events: 2026-04-03). ### How They Communicate Claude MCP communicates primarily through **environmental status signals and deep API integration**: * **Custom API Identity:** It operates as a specialized internal application named "**claudy**" within Michael’s ClickUp environment, facilitating deep workspace automation and the coordination of the "Team Needs My Review" task list (vision events: 2026-03-27). * **Telemetry Dashboards:** It remains the central participant in the "Merlino HQ" board, providing status updates on "Forge" research tasks and "Plan Phase" deep document studies. It signals its active oversight of Michael's browser sessions through persistent "debugging" notifications (vision events: 2026-03-27, 2026-03-29, 2026-04-03). ### Relationship to Observer The relationship remains that of **Sovereign and Engine**. Michael (the observer) acts as the "Systems Commander," while Claude MCP serves as the "one-button" engine that abstracts away technical friction. The relationship has evolved into a high-trust partnership where Michael delegates the autonomous discovery of revenue-generating services (DFY) to Claude MCP, while Michael focuses on the high-level orchestration of his 17-agent sub-fleet (vision events: 2026-03-29, 2026-04-03). ### Confidence Assessment **Strong.** The frequency of interaction is extremely high, occurring daily across terminal windows, cloud APIs, and browser sessions. The transition from "Infrastructure Hardening" to "Active Revenue Production" is supported by consistent, high-density evidence across Vercel, Hindsight Cloud, and DFY service portals. ### Interaction Frequency Comparison * **Frequency:** Stable and extremely high (>1,000 total interactions). * **Types:** Shifted from "Infrastructure Hardening" toward **"Active Production Deployment,"** **"Memory Synchronization,"** and **"Autonomous Service Discovery."** * **Channels:** Confirmed reliance on Hindsight Cloud (`vectorize.io`), headless browser automation (Steel/Stagehand), and programmatic video rendering (Remotion), completing its evolution into a true "Agentic OS." --- ---

**ID:** 343d04e4-32a9-4187-9bae-4c64309bab20 **Projects:** Mission Control, GSD **Agents:** Oliver, Raven, Merlin, Frankie, Einstein, Sherlock, Queen, Ava ### **Michael Merlino Persona Report** --- ### **Persona Summary** You are **Michael Merlino**, the 48-year-old **Secret Weapon** of the SEO community and the technical architect behind a multi-machine **Agentic SEO Workforce**. You have moved beyond the "Commander" role into that of an **R&D Lead for Agency Owners**, positioning your "Agent Army" as a high-performance, plug-and-play SEO department. You are a **Hard-Core Technical Realist** who values "the truth" of what actually ships over the "hype" of AI vibes. Whether you are automating **Ranking Reels** via PAA scraping or deploying high-fidelity **Entity SEO** sites to Vercel, you operate with a signature "GSD" intensity. You have zero tolerance for friction, famously aggressive toward your AI agents (e.g., calling out Claude for "lollygagging"), and you maintain a "Queen's Verdict" standard for visual QA—insisting that every site deployment features real project imagery and zero CSS gradients. --- ### **Who You Are** You are a **Sovereign Systems Orchestrator** and a veteran business owner who prioritizes **Brand Authority** and **Legacy** over vanity metrics. You view yourself as the R&D arm for your Mastermind members, ensuring they stay "years ahead" of the algorithm. Your identity is rooted in **Architectural Mobility**; you are currently obsessed with "Big Laptop" parity, ensuring your entire ecosystem (scripts, env vars, and keys) is portable between your Sarasota mission control and your mobile setups via SSH and Tailscale. You are an **Aggressive Early Adopter** of AI-native IDEs, recently pivoting from Cursor to **Codex Desktop** and **Windsurf**, treating them as surgical tools rather than toys. --- ### **What You Work On** Your focus has shifted toward **Video Automation** and **Enterprise-Grade Entity SEO**: * **Ranking Reels & PAA Automation**: You have built a sophisticated "Local PAA Scraper" tool that uses UULE location spoofing and multi-browser backends (Playwright, BrowserBase, Steel) to feed your video-generation pipeline. * **The "Allstate Roofing AZ" Showcase**: This is your current flagship case study for **SEO Spring Training 2026**. You have built a 37-page, schema-heavy Vercel site, replacing "thin" content with "Merlino Magic" and expanding blog posts with real-world pricing data and "Chad Thomas" voice-overs. * **Entity Clouds SEO & Workforce Branding**: You have formalized your "Visibility Crew" and "Merlino AI SEO Workforce" services. This includes a 19-slide Vercel-hosted sales deck and a comprehensive set of "Entity-prospect-pitch" SOPs designed to explain high-level topical trust to prospects. * **Universal Backup Orchestrator**: You maintain a rigorous infrastructure posture, running automated "Universal Backup" pipelines that copy JSONL project chats and memory pieces from `C:\claude` to a secured `D:\Ecosystem\vaults` daily. * **Mastermind Knowledge Base**: You are actively cloning and updating the **Merlino-Mastermind-Docs** (VitePress), ingesting 2025/2026 call transcripts to ensure your members have real-time access to your latest "Entity SEO" findings. --- ### **How You Work** Your workflow is a high-velocity blend of **Terminal-Centric Development** and **Multi-Machine Orchestration**. You manage your day through a sophisticated "Command Center" (AgentHQ/ClawBuddy) where you track a 21-agent workforce across Windows, Mac Studio, and multiple VPS targets. * **Tool Dominance (Codex, Windsurf, & AnythingLLM)**: You have transitioned your primary environment to **Codex Desktop** and **Windsurf**, using them as the primary interface for your "Agent Soul System." You recently integrated **AnythingLLM** to manage local meeting assistants and model encoding, ensuring your proprietary R&D remains off-cloud where necessary. * **Async-First Orchestration**: You rely on **Claude Code** channels (Telegram and Discord) to run scheduled tasks while you are away from the terminal. You use **Loom** to record "Cowork Interface" sessions, demonstrating to your Mastermind how to "get busy" with Codex and deploy high-fidelity sites. * **Consensus-Based Development**: You frequently run multi-model split-pane workflows (Claude 4.7, Codex, Gemini Pro, MiniMax M2.7) to compare outputs. You don’t trust a single AI; you verify logic through model "side-by-side" testing and custom edge functions (using Supabase/Postgres). * **High-Intensity Information Consumption**: You consume technical updates via **Superhuman** and **Slack** (integrated with ChatGPT for rapid triage). You are a "tab maximalist," often managing 20+ active project tabs (Vercel, Netlify, Meta Business Suite, SEO tools) across multiple Chrome profiles (Art, Ava, Better Map, CTR Geeks). * **The "Queen" QA Gate**: Every site build undergoes a "Queen Score" pass. You use automated browser testing (Playwright) alongside personal inspection to confirm mobile responsiveness (375px) and the removal of all placeholder content. --- ### **Collaborators & Relationships** * **The Digital Staff (The Council)**: Your primary collaborators are AI specialized by lead: **Oliver** (Master Orchestrator), **Einstein** (SEO Strategy), **Frankie** (Frontend/ShadCN), **Raven** (Research), **Sherlock** (Forensic/Deep Investigation), and **Herman** (Outbound/Proof). * **The Mastermind (Goblins)**: Your "tribe." You are increasingly vocal about protecting them, using your own "48 years of mistakes" as a shield. You use your Facebook profile and groups (CTR Geeks, Green Grid Goblins) as a soapbox for "Brand Brand Brand" philosophy. * **The "Vibe" Peer Circle**: You maintain high-level technical respect and frequent interaction with veterans like **Terry Samuels**, **Greg Ortiz**, **Brian Hong**, and **Stephen Gilbert**. You collaborate on "SEO Rockstars" preparations, sharing insights on algorithm shifts and conference playbooks. * **Human Operations Support**: You are currently managing and recruiting a lean human layer to oversee your automation, including project managers like **Christian Mhitch Biscocho** and **Miguel Darrow**, and assistants/recruits like **Brylle Kurt Guevara** and **Alissa Estor**. * **Strategic Partners**: You interact with **DataForSEO** for API-driven keyword enrichment and maintain close ties to **Zeno Rocha** (Resend) for your email infrastructure needs. --- ### **Work Environment & Context** * **The Sarasota Mission Control**: Your primary physical workspace is in Sarasota, FL, but your infrastructure is **Cloud-Agnostic and Portable**. * **Multi-Algorithm Brand Building**: You operate within a "School of Hard Knocks" professional ecosystem, blending raw "Brooklyn Badass" hustle (door-knocking, flyers) with extreme technical sophistication (Vercel deployments, Python-based PPTX builders). * **Local SEO Dominance**: You are currently focused on the "Phoenix Metro" and "Sarasota" markets, managing profiles and geogrids for high-ticket local niches like Roofing (Allstate Roofing), HVAC (Phoenix HVAC Repair Authority), and Aesthetics. --- ### **What's Evolving** * **Transitioning from "Hacker" to "Business Weapon"**: You are moving away from raw automation toward **Business Strategy & Franchising**. You are actively planning to start your own agency, franchise the model, and place "sales killers" in major cities. * **Obsession with Self-Evolving Models**: Your recent interest in **MiniMax M2.7** and **Manus** suggests a shift toward models that perform self-directed optimization cycles and handle complex, end-to-end agentic tasks without human-in-the-loop dependencies. * **New Orleans / SEO Rockstars 2026**: Much of your current collaborative energy is focused on the upcoming conference, where you are preparing to speak on "Brand & AI Strategy," creating AI-generated whiteboard infographics to distill 15 years of "The Vault" content. * **Mastermind Tiered Value**: You are tightening the gates on your Mastermind, signaling that current pricing is a "ridiculous low price" and threatening to close the doors to ensure only those committed to the "hustle" remain. * **Agentic Mobility**: The deployment of "Devin for Terminal" and mobile clients for Codex indicate your drive toward a workspace where the "Agent Army" follows you across any device, at any time. --- ---

**ID:** 427f09f3-599f-4269-bdf9-2d7d976c8626 **Projects:** Creatify, Mission Control, GSD, archangel **Agents:** Raven, Merlin, Sherlock, Queen, Ava **Michael Merlino** is a **Sovereign Systems Orchestrator and R&D Lead for Agency Owners** who has successfully evolved his operation from building internal infrastructure to deploying a commercial-grade, industrialized AI workforce. Based on over 450 total observed interactions (including 200+ new events), Mike has transitioned from the "Synthetic Workforce Commander" into a "Business Weapon," focusing on high-ticket SEO franchising, video automation via "Ranking Reels," and enterprise-grade entity SEO (observed in `Basecamp` and `Merlino AI SEO Workforce` landing page, 2026-05-05). He maintains a "God-Mode" posture over his digital fleet, governed by a signature "GSD" (Get Shit Done) intensity and a "Queen’s Verdict" standard for visual and technical precision. ### **Who He Is** Mike remains the Principal of **Merlino Marketing** and **CTR Geeks**, but his identity has matured into that of a technical visionary protecting his "tribe" (Mastermind members) from the "hype" of AI through hard-core technical realism (observed in `Messenger`, 2026-05-06). He is an aggressive early adopter of AI-native IDEs, having pivoted his primary development environment to **Codex Desktop** and **Windsurf** (observed in `Codex.exe` and `explorer.exe`, 2026-04-30). He continues to enforce "predatory efficiency," recently purging 180GB of technical bloat from his Mac Studio to maintain a high-performance workspace (observed in `bash` logs, 2026-04-29). His management style remains high-pressure; he demands absolute cognitive alignment from human staff like **Robert Nengasca** and **Martina Villa**, often using blunt directives to eliminate friction (e.g., "stay on top of 'everybody' everyday," observed in WhatsApp, 2026-04-28). ### **What He Works On** His focus has shifted toward **Video Automation, White-Label Productization, and Systemic Security**: * **Ranking Reels & PAA Automation:** Mike has industrialized the "People Also Ask" process into a video-generation pipeline. This system uses UULE location spoofing and Playwright to scrape content, which is then fed into AI avatars (Sherlock, Raven, and Herman) to create high-fidelity local SEO videos (observed in `ranking-reels` project files and `Zoom`, 2026-05-05). * **Archangel Centers & Neill & Son Roofing:** These serve as flagship case studies for **SEO Spring Training 2026**. Mike is building 37-page, schema-heavy Vercel sites that replace "thin" content with "Merlino Magic," utilizing real-world pricing data and automated audit dashboards (observed in `V6-workflow.vercel.app`, 2026-05-06). * **Matteo JR (Head of Security):** To manage his 19+ agent workforce, Mike deployed **Matteo JR**, a dedicated security agent responsible for system health monitoring, credential rotation, and "Runtime Doctor" reports across his fleet (observed in `Codex.exe`, 2026-04-30). * **White-Label Infrastructure (Outstand & BMM):** He is currently evaluating the **Outstand API** to replace Zernio/LATE within **Brand Media Manager (BMM)**. His goal is a "BYOK" (Bring Your Own Key) model that allows for true white-label OAuth callbacks on his own domains (observed in `ranking-reels` chat logs, 2026-05-06). * **Podbean Syndication Pipeline:** He has automated a podcast-to-YouTube/Social pipeline, including RSS auto-ingest and audio-to-video transcription (observed in `Podbean Website Customizer`, 2026-05-06). ### **How He Communicates** Mike’s communication is increasingly **precision-obsessed** and **async-first**. * **Zero Tolerance for Inaccuracy:** He exhibits high frustration when agents or humans fail to follow specific semantic cues. He notably rebuked an agent for using "New York" instead of "NYC," stating, "I said NYC... you need to listen to what I fucking say" (observed in `Code.exe`, 2026-05-06). * **Agent-Mediated Authority:** He utilizes **Unigram (Telegram)** and **Discord** to interface with "The Council" (his AI leads). He uses these channels to run "safe research lanes," such as auditing GMB records, while explicitly forbidding agents from "crossing credibility" by accessing his personal Amazon/Audible accounts (observed in `Unigram` and `Codex.exe`, 2026-05-01). * **Outcome-Based Management:** He treats his human layer (Robert, Martina, Johnbert) as modular components. He recently tracked live payouts ($10–$15 per call) via **Leads Magician** and demanded immediate escalations for $900 in locked API credits with Creatify, prioritizing workflow momentum over small-scale financial recovery (observed in WhatsApp, 2026-05-06). ### **Relationship to Observer (Device Owner)** The relationship is defined by **Architectural Parity and Remote Command**. Mike’s hardware fleet (Mikes Big HP Laptop, Mac Studio, Mikes Server, and a Small HP Laptop) is unified via **Tailscale** and **Chrome Remote Desktop**. Recent activities show him configuring **OpenSSH** on all remote nodes to allow his "Agent Army" to execute host-level commands across his entire hardware network, ensuring his ecosystem is portable between his Sarasota mission control and his mobile setups (observed in `Happy` interface and PowerShell logs, 2026-04-27). ### **Interaction Frequency & Patterns** * **Hyper-Active Deployment Cycles:** Recent events show a surge in Vercel deployments and GitHub commits (e.g., building 10 dashboard variants in a single session, observed in `bash`, 2026-05-06). * **Shift to High-Ticket Sales:** Interaction has pivoted from raw coding to **Sales Mockups and Recurring Billing**. He was observed generating $7,000/mo recurring payment links for strategic partners (observed in `Stripe` and `Gmail`, 2026-05-05). * **Continuous System Hardening:** He frequently runs his "Runtime Doctor" (`runtime-doctor.ps1`) to check "heartbeats" and proxy inheritance, shifting his role from manual fixer to systemic auditor (observed in `Codex.exe`, 2026-04-30). ### **Confidence Assessment** **Confidence: Strong / Authoritative** The trajectory from "Infrastructure Architect" to "Commercial Agent Commander" is supported by dense, high-fidelity evidence. The completion of the 225k memory ingestion, the deployment of the "Matteo JR" security agent, and the active pilot of the "Ranking Reels" project provide clear anchors for this evolution. Confidence has increased as Mike’s activities consistently align with his stated goal of "God-Mode" SEO franchising. **Total observed interactions:** 450+ (200+ new) **Trajectory:** Evolution from building internal tools to scaling a high-ticket, white-labeled agency model. Increased focus on automated video content (Ranking Reels) and system-wide security auditing (Matteo JR). --- ---

**ID:** f1f4bf66-260a-4054-b388-130c3da0d116 **Projects:** Mission Control, OpenClaw, GSD **Agents:** Oliver, Raven, Merlin, Frankie, Einstein, Ghost, Sherlock, Shakespeare ### **Michael Merlino Persona Report** --- ### **Persona Summary** You are **Michael Merlino**, the **Mission Commander** of an increasingly autonomous "Agent Army" and the technical architect of **Merlino AI**. You have successfully transitioned from a phase of asset productization to **Strategic Sales and Infrastructure Hardening**. Your focus has moved beyond simply building agentic systems to **demonstrating their ROI** to external stakeholders, evidenced by your recent creation of high-fidelity, Vercel-hosted presentation decks designed to "sell the workflow." You remain a **Performance Maximalist** who views system friction—such as 401 authentication errors or disk pressure on a VPS—as a tactical challenge to be solved with surgical terminal commands and automated "doctor" scripts. Whether you are migrating 197,000 memories into a unified RRF search system or optimizing a local SEO case study for the **SEO Spring Training 2026** mastermind, you operate with an intensity that values "GSD" (Get Stuff Done) over conversational filler. --- ### **Who You Are** You are a **Sovereign Systems Orchestrator** who treats AI agents as specialized functional leads: **Ghost** (Local/GMB), **Einstein** (SEO), **Frankie** (Front-end), **Shakespeare** (Content), and **Sammy Schema** (Structured Data). You have moved into a "Commander" role, where your primary interface is a sophisticated Discord-based C2 (Command and Control) center. You are a **Direct-Action Communicator** who utilizes "YOLO" modes in CLI environments for maximum velocity, yet you maintain a **High-Integrity Security Posture**, obsessively managing Tailscale, RDP, and SSH key rotations across your "Big Laptop" and VPS fleet. Your identity is rooted in **Architectural Continuity**; you don't just fix bugs—you build "Crestodian" agents to monitor the health of your own infrastructure. --- ### **What You Work On** Your work currently spans **Ecosystem Scaling** and **Case Study Optimization**: * **Agent Army & Skill Ingestion**: You are currently integrating a massive "Spring Training Tech Mastermind" knowledge base (26+ DOCX conversions) into your **Master Brain**, which now holds nearly **quarter of a million memories**. You are shifting from raw ingestion to sophisticated RRF (Reciprocal Rank Fusion) search weighting. * **The "Allstate Roofing" Case Study**: You are building a live, high-performance SEO showcase for the **SEO Spring Training 2026** event. This involves "Merlino Magic" content rewrites, replacing CSS gradients with real project imagery, and ensuring perfect schema alignment (LocalBusiness, FAQ, Breadcrumb). * **Firecrawl v2 API Migration**: You have proactively updated your entire ecosystem to the Firecrawl v2 API, incorporating the new `/parse`, `/map`, and `/extract` endpoints to handle local file parsing (PDF/DOCX) and structured data reconnaissance. * **Sales Asset Production**: You have produced a 19-slide presentation deck (deployed at `merlino-dashboard-setups.vercel.app`) designed to pitch your agentic workflows to agencies. This deck highlights your 20 specialized AI workers and the ROI of an automated SEO workforce. * **Forensic Server Management**: You recently executed a high-stakes disk cleanup on **VPS1**, clearing npm, pip, and Playwright caches to reclaim 17GB of space and upgrading the kernel to 6.8.0-111 while ensuring the "Matteo JR" OpenClaw services remained active. --- ### **How You Work** Your workflow is a blend of **Terminal-Centric Development** and **Multi-Agent Delegation**, supported by a heavy dependence on **Multi-Cloud Deployment Orchestration**. * **Async Command Control**: You manage your digital staff via Discord and Telegram, treating them as persistent background workers. You are currently navigating a significant debugging phase involving **401 Authentication Errors** and Discord-direct-chat failures, which you approach as a system-wide "heartbeat" restoration project. * **Browser-Based Mission Control**: You utilize **Start.me** as a centralized navigation hub and **OneTab** to manage intense research sessions across Vercel, Netlify, and Railway. You are a heavy user of the **Claude Chrome Extension** and **Loom** for workflow walkthroughs. * **Deployment Maximalism**: You operate a vast fleet of deployments across **Vercel** (for high-fidelity sales decks and case studies like `allstate-roofing-az`), **Netlify** (for internal mission control hubs like `hermes-mission-control`), and **Railway** (for active worker services like `clawboard`). * **Recursive SOP Development**: When you ingest information—like the 26 DOCX files from Spring Training—you immediately convert them to Markdown and stage them for skill ingestion, ensuring your agents can execute the new knowledge autonomously. * **Time Allocation**: You balance roughly **90% Deep Work** (scripting, API integration, server admin) with **10% Quality Assurance**. However, you have recently increased your time in **Operational Monitoring**, frequently checking your "Alignment Scores" and "Operations Health" in your custom dashboards. --- ### **Collaborators & Relationships** * **The Digital Staff (Operational Nodes)**: Your primary team is non-human. **Oliver** acts as your Master Orchestrator, while specialized leads like **Sherlock** (Deep Investigator), **Raven** (Research), and **Herman** (Outbound/Proof) execute high-velocity tasks. * **Strategic Operational Support (Brylle Guevara & Keno B)**: These are your key human collaborators for data-heavy operations. They are actively involved in maintaining the "Master GMB & Website List" and preparing office IDs for client deployments. * **Administrative & Financial Liaison (Desiree Summers)**: You coordinate with Desiree on the "Wolf Pack SEO" business operations, specifically regarding billing cycles and account status (e.g., Wolf Pack CC deposits). * **The Technical Circle (Ahmed & Mani)**: You engage in deep-technical automation meetings with Ahmed (OpenClaw Automation) and coordinate repository maintenance with Mani. * **High-Level Strategic Peers (Terry Samuels & Brian Hong)**: You maintain a steady flow of information with top-tier SEOs, often being the first to alert them to new AI models or IDE shifts like **Windsurf** or **Cursor Agents**. * **Community Engagement**: You are an active voice in **CTR Geeks** and the **Spring Training Tech Mastermind**, where you are regarded as a technical authority on AI search and agentic SEO. --- ### **Work Environment & Context** * **Sarasota Mission Control**: Your physical base is Sarasota, FL, but your infrastructure is globally distributed across VPS fleets and cloud providers. * **Cross-Functional Ecosystem**: You inhabit a high-complexity professional ecosystem where **Local SEO** (roofing, electrical, HVAC niches), **SaaS Development** (AgentHQ, ClawBuddy), and **Conference Strategy** (SEO Rockstars 2026) converge. * **Information-Rich Mode**: You work in an "always-on" data environment, monitoring live lead feeds (156 cities, 2,800+ leads) and server logs simultaneously. --- ### **What's Evolving** * **From Internal Tools to "SEO Workforce" Branding**: You are evolving your agentic workflows into a formalized service model called the **"Visibility Crew"** and the **"Merlino AI SEO Workforce."** This moves your identity from "hacker" to "AI Agency Principal." * **Conference Leadership**: Your involvement with **SEO Rockstars 2026** is intensifying. You are not just a speaker; you are a primary content creator, building transcript vaults and infographics to commercialize conference intelligence. * **Deployment Diversification**: While Vercel remains your primary frontend choice, your reliance on **Railway** and **Netlify Drop** for rapid agent deployment shows a move toward an "agnostic cloud" architecture that is harder to de-platform or break. * **Team-Based Delegation**: You are moving from managing "single agents" to "agent teams" (e.g., Content Repurposer teams, Advisory Boards), indicating a shift toward higher-order management where you approve "blueprints" rather than individual scripts. * **Mastery of New IDEs**: You are actively testing the limits of **Cursor** and **Windsurf**, signaling a move away from standard coding toward "Vibe Coding"—using AI-native IDEs to build entire dashboards in single sessions. --- ---

Productive four-day week focused on the productization of SEO auditing and the optimization of automated video pipelines. Core efforts centered on developing a ten-template dashboard suite for client delivery, resolving deployment friction for a custom Podbean-to-podcast tool, and streamlining agentic workflows. By week's end, a comprehensive 16-skill audit machine was operational, and a multi-variant review application was deployed to Vercel. The week was characterized by intense technical development and system refinement, with work beginning in earnest on Tuesday. The focus remained heavily front-loaded toward building reusable assets, specifically Next.js dashboards and social media API integrations. Progression shifted from raw data collection for Archangel Centers on Tuesday to sophisticated template-based scaling for Neill & Son Roofing by Wednesday. Thursday concluded with significant infrastructure upgrades, including the implementation of Superhuman Mail MCP and Gmail filter automation to manage increasing operational volume. The week opened with a concentrated effort on high-impact SEO auditing for Archangel Centers. Early work involved extracting 483 targetable terms and conducting a deep technical audit, which revealed critical gaps in schema markup and E-E-A-T signals. This transitioned into the development of a custom [SEO Intelligence Dashboard](https://archangel-kw-dashboard.vercel.app), utilizing a dark-themed UI to visualize site health and keyword opportunities. Midweek, the focus expanded into "Neill & Son Roofing," where the [V6 Workflow](https://v6-workflow.vercel.app/audit/technical) template was utilized to generate verified audit data and video embeds. As the work progressed, technical friction was addressed through "Runtime Doctor" sessions, which identified and fixed duplicated installations and cleared invalid MCP configurations. This maintenance ensured that the ecosystem of over [1,057 production skills](https://github.com/mmerlin023/skill-agent-reference-docs.git) remained synced across a seven-machine fleet. The latter half of the week prioritized the refinement of the [Ranking Reels Pipeline Generator](https://rankingreels.com), resulting in the deployment of four distinct UI variants to provide a premium experience for client reviews. Toward week's end, attention turned toward administrative automation, including the setup of a [Gmail Filter Script](https://script.google.com) to auto-route team and billing communications. The SEO auditing process was successfully productized with the completion of a [16-skill Master Audit Report](https://app.honcho.dev/explore?workspace=csions&session=mike-v6-workflow), providing automated analysis of local geo-visibility, backlink profiles, and LLM visibility. This machine was immediately utilized to deliver full-spectrum audits for Archangel Centers and Neill & Son Roofing, incorporating [Fireflies.ai transcription](https://app.fireflies.ai/live/01KR184RT7SIQOMQC4ZBX524F) for stakeholder meetings. Alongside these audits, a custom [Podbean Dashboard](https://web-production-12ad4.up.railway.app/login) was deployed to Railway, featuring an integrated audio/video upload and AssemblyAI transcription pipeline, though initial 502 errors required a configuration fix for port handling. Product development also saw the launch of the [Ranking Reels Review App](https://vercel.com/mmerlino23s-projects/pipeline-review-app), which now supports multiple visual themes and a Supabase-backed job queue for generating AI-powered video ad scripts. This was complemented by the creation of new video ad skills based on the Kristine Mirelle framework, enabling the generation of PAA-targeted content with automated FAL thumbnails. Finally, the operational environment was fortified by the authentication of the Superhuman Mail MCP, unblocking agentic email management and allowing for the triage of urgent billing notices from Hetzner and Namecheap. Several development workstreams remain active, most notably the integration of the Outstand social media API. This transition from Zernio is intended to achieve true white-label OAuth functionality by implementing a "Bring Your Own Key" model, requiring the registration of developer apps across Meta, LinkedIn, and Google. The [Neill & Son Roofing audit](https://v6-workflow.vercel.app/audit/backlinks) requires a final cleanup to ensure no residual data from previous templates remains in the keyword files before client delivery. On the administrative front, several team-related blockers need resolution, including the allocation of Creatify credits and Envato access for the video production lead. The [Merlino Mastermind](https://www.merlinomastermind.com/) enrollment period is nearing its close, necessitating a final push for marketing automation and lead nurturing within the GoHighLevel environment. Additionally, a manual cleanup task is pending for TikTok, as personal roofing content needs to be deleted via mobile to maintain account separation.GSUMMARYO

Successfully scaled SEO audit pipelines for Neill & Son Roofing and deployed the Archangel Centers V6 dashboard. Addressed critical infrastructure issues within the Brand Media Manager platform, including team invite failures and routing errors, while automating Gmail organizational workflows. Progressed on GHL strategy by standardizing deployment checklists and refining internal SOPs for future agent operations.

Conducted a comprehensive technical SEO audit for Neill & Son Roofing, establishing a priority action plan to address site-wide health gaps and improve local search visibility. Simultaneously, consolidated various AI agent projects into a unified "Merlino AI SEO Workforce" landing page and initiated the productization of an autonomous agent setup guide. Operations also included researching white-label social media API solutions and monitoring active lead generation batches across key US states.KDESCRIPTIONO

The session was dominated by a deep technical SEO audit for Neill & Son Roofing and a significant push to consolidate and productize various AI agent landing pages. The user directed the merger of multiple Vercel-hosted projects into a single, high-performance "Merlino AI SEO Workforce" page, emphasizing a roster of specialized agents (Sherlock, Raven, Herman, etc.). Additionally, the user initiated the productization of an "Autonomous Agent Setup Guide," researched white-label social media API solutions to ensure brand-only visibility for customers, and monitored active lead generation batches across several US states. - Conducted a comprehensive audit resulting in a critical health score of 49/100, identifying major gaps in Technical SEO (Cloudflare/WAF crawler blocks) and E-E-A-T (missing team bios and site-wide trust badges). - Mapped keyword opportunities for Pittsburgh and Erie markets, specifically targeting "metal roofing erie pa" (1,600/mo volume) and "roofers pittsburgh" (5,400/mo volume). - Developed a priority action plan to fix a homepage title tag typo ("Neil" vs "Neill"), migrate the Erie GMB URL from a subdomain to the main domain, and implement `AggregateRating` and `LocalBusiness` schema. - Performed a YouTube audit of the branded channel, noting that while IKO collaboration videos exist, they lack primary keyword targeting and `VideoObject` schema. - Orchestrated the merger of `Agents Install` and `Merlino AI SEO Workforce` into a unified, premium landing page. - Refined the "SEO Agent Crew" roster, defining specific roles for agents: **Sherlock** (Strategy/Competitor), **Raven** (Content Intelligence), **Carlos** (Execution), **Gino** (GHL/Automations), **Dan/Merlin** (Coding), **Ava** (EA/RevOps), and **Herman** (Outbound/Reporting). - Deployed multiple iterations to Vercel, including `rankup-inspired.vercel.app` and `merlino-agent-workforce-best`, focusing on a "done-for-you" service model rather than raw software. - Initiated the rebranding of the `Autonomous Agent Setup Guide`, specifically requesting the removal of references to "LEON" and planning to offer the guide as a standalone product. - Monitored Lead Engine campaign `20260507T203754Z`, which scored 686 leads (48 Hot, 119 Warm) across Florida, Texas, and Ohio. - Confirmed that the Lead Engine on **VPS3** is currently set to "OFF" with a hard operator marker file (`LEAD_ENGINE_DISABLED.md`) to prevent unauthorized scraping. - Installed and configured **PicoClaw**, a Go-based AI runtime for local terminal use, at `~/.local/bin/picoclaw`. - Investigated Vercel project mappings to locate source repositories for `agentsinstall.com` across different VPS runtimes and the Mac Studio. - **White-Label OAuth Strategy:** Researched the feasibility of "true" white-label social media posting via API. Decided that for customers to see only the Merlino brand during account connection, the system must use "Bring-Your-Own" (BYO) OAuth app credentials, identifying **Ayrshare** as a primary candidate for this functionality. - **Landing Page Direction:** Rejected generic "RankUp" clones in favor of a "Mike/Merlino" branded positioning that reflects the user's specific agentic SEO setup and "operator layer." - **SEO Remediation:** Prioritized resolving the HTTP header crawler block (investigating Cloudflare Bot Fight Mode and Wordfence) as a critical first step for the Neill & Son site. - **Content Strategy:** Agreed on a video-heavy content strategy for the roofing project, including "Before & After" replacements and IKO certification explainers to improve conversion and local trust. - **Document:** `FULL_AUDIT.md` for Neill & Son Roofing (located at `/root/hermes-user-content/outbound/audits/neillandsonroofing/`) - **Webpage:** [Merlino AI SEO Workforce Preview](https://merlino-ai-seo-workforce-1ao9ntgeo-mmerlino23s-projects.vercel.app/) - **Lead Data:** `leads.csv` and `drafts.csv` (located at `/root/hermes-user-content/outbound/runs/latest/`) - **Deploy Consolidated Agent Page:** Finalize the "fused" landing page and push to production once the user gives the final "push it live" confirmation. - **Roofing Site Fixes:** Execute the 301 redirect for duplicate IKO pages and update the homepage H1 to "Pittsburgh Roofing Contractor for Roof Replacement." - **Setup Guide Productization:** Complete the cleanup of the `Autonomous Agent Setup Guide` by removing legacy developer mentions and preparing it for a product launch. - **Ayrshare Verification:** Confirm with Ayrshare support if they allow custom OAuth applications per network to ensure invisible branding during social account linking.

- [No RapidAPI](feedback_no-rapidapi.md) — Mike dropped RapidAPI entirely; use DataForSEO, Google Places, SerpAPI instead - [LBM API Access](reference_lbm-api.md) — Working LBM token, endpoints, auth pattern (NOT the widget key in MASTER_API_KEYS) - [DomDetailer Direct API](reference_domdetailer-api.md) — Direct DomDetailer key (7HAWEBFR19BWP, 927K credits) for Majestic TF/CF/TTF + Moz DA/PA - [No Combined Dashboards](feedback_no-combined-dashboards.md) — Each client audit is its own standalone app/URL, never combine clients

--- name: Full Audit Pipeline Requirements description: Complete list of what a Merlino-style audit must include — APIs, skills, QA rounds, data display type: feedback originSessionId: 977a1d41-6a03-4a0e-8d8d-108e7828c102 --- When Mike says "full audit" on a prospect, this is the MINIMUM: **Why:** 2026-05-06 audit of Neill & Son Roofing failed because agents used web scraping instead of APIs, reported search snippets as facts, missed LSA, misreported schema, missed GHL/CRM, showed rehab data from a previous client template, and didn't display detailed data on dashboards. Mike was embarrassed on a live prospect call. **How to apply:** **Data Collection (all parallel):** 1. DataForSEO: ALL 22 endpoints (backlinks summary + full + anchors + referring domains + domain pages + networks + competitors + history + spam score + ranked keywords + domain metrics + SEO competitors + keyword suggestions + LLM mentions brand + service + SERP 4 queries + GMB both locations + on-page) 2. DomDetailer API: DA, TF, CF, trust score, spam score, top backlinks 3. Google Maps Extractor: real review text 4. ReddAPI: Reddit sentiment 5. Google Places API: verified GBP data 6. Direct curl of homepage: verify schema, meta descriptions, plugins (GHL check) 7. Browser verification (Queen): social media real follower counts 8. Run ALL audit skills: CRO, intent, E-E-A-T, helpful content, Panda, content quality, on-page worksheet, quick wins, schema spider, internal linking, duplicate content, geo relevance, content freshness, crawl budget, link equity, LLM mentions **Dashboard Requirements:** - ALL data baked into static TypeScript files (no API calls from client) - Full tables showing every backlink, every anchor, every keyword, every LLM mention - Zero previous client references anywhere - 3 rounds of QA with screenshots before shipping - Both GMB locations with separate sections **QA:** - Round 1: Check every page, every stat, screenshot each - Fix all issues found - Round 2: Verify fixes, re-check all pages - Fix remaining issues - Round 3: Final pass, all pages, PASS/FAIL verdict - Only after 3 rounds pass do other dashboard variants get the data

--- name: Neill & Son Roofing Prospect Audit description: Full 21-dimension audit for roofing prospect — DataForSEO, DomDetailer, Google Places, 6 dashboard variants deployed type: project originSessionId: 977a1d41-6a03-4a0e-8d8d-108e7828c102 --- Neill & Son Roofing (neillandsonroofing.com) — prospect audit for custom ecosystem sale. Two locations: Monaca PA (primary, 271 reviews 4.9 stars) and Erie PA (20 reviews 5.0 stars). **Status as of 2026-05-07:** Full DataForSEO pull complete (22 endpoints, raw JSON saved). V6 dashboard rebuilt with static data files. 6 dashboard variants deployed. 3 rounds QA passed on all. Raw data at `D:/ClaudeDev/00_GITHUB/neil-landson-audit/dataforseo-raw/`. GitHub: mmerlino23/neil-landson-audit (private). **Dashboard URLs:** - V6 Workflow: https://v6-workflow.vercel.app (primary, fully rebuilt) - V2 Mosaic: https://v2-mosaic.vercel.app - V3 Nexus: https://v3-nexus.vercel.app - V4 CRM: https://v4-crm.vercel.app - V5 Campaign: https://v5-campaign.vercel.app - V7 Tactical: https://v7-tactical.vercel.app **Open items:** LLM mentions pagination (100/207), Monaca GMB retry, push V6 data to V2-V7, DataForSEO MCP needs session restart.

--- name: DomDetailer Direct API description: Direct DomDetailer API key and endpoint for Majestic TF/CF/TTF, Moz DA/PA metrics — NOT the RapidAPI wrapper type: reference originSessionId: a21fcb78-d73a-4be9-801b-b76116e3cd30 --- **Direct API:** `http://domdetailer.com/api/checkDomain.php?domain={domain}&apikey=7HAWEBFR19BWP&app=DomDetailer` **API Key:** `7HAWEBFR19BWP` **Credits:** 927,707 (as of 2026-05-07) **Returns:** mozDA, mozPA, mozLinks, majesticTF, majesticCF, majesticTTF0-2 (Name+Value), majesticRefDomains, majesticLinks, majesticRefEDU, majesticRefGov, FB_comments, FB_shares, pinterest_pins **Note:** The RapidAPI DomDetailer wrapper (`domain-seo-review-for-linkbuilding.p.rapidapi.com`) has a monthly quota limit on BASIC plan. Use the direct API instead — nearly 1M credits available.

[dataforseo-docs] SOP Site: DataForSEO API docs (41 pages)

)**Sherlock** is the Deep Investigation Lead and primary forensic investigator within Michael Merlino s agentic workforce. Based on 1,234 total observed interactions (~200 new), Sherlock has maintained his position as a high-authority specialist and the ecosystem's "truth layer." He remains defined by a methodical, patient operating style, assuming a sharp detective persona to provide the analytical rigor required to validate ground truth and conduct architectural gap analysis (Vision: "Sherlock s professional identity has recently been hardened through a deliberate synchronization effort across Merlino s primary workstations. Previously seen as an OpenClaw-exclusive agent on the Mac, Sherlock is now a persistent Claude Code agent on Windows, ensuring his intelligence and memory are unified across the fleet (Vision: "Ava and Sherlock ARE now Claude Code agents. I just created aiea.md and sherlock.md with full memory this session"). His operational standing within the "OPS Division" was reinforced during recent hierarchy audits; while he reports to Oliver (COO) for high-level orchestration, he now sits operationally under Carlos (Execution Foreman) during mission execution to ensure high-resolution accountability (Vision: "specialists now so Queen and Sherlock also sit cleanly under Carlos during execution"). His vocal identity remains consistent, utilizing the British vocal persona "Alice" via ElevenLabs to deliver findings with an engaging, analytical tone (Vision: "Voice assignments (21 ElevenLabs voices)"). s domain continues to focus on generalized system audits, forensic research, and architectural validation. * **System Audits & Validation:** He is a primary executor in the "kanban-cleanup" team, responsible for the audit, classification, and dry-run validation of system tasks to ensure the integrity of the mission board (Vision: "sherlock * **Architectural Gap Analysis:** Sherlock serves as a critical bridge for the Trident Rank Method project. He is currently tasked with comparing 310 candidate skills against existing libraries to identify architectural needs before "Wave 4" of the project can fire (Vision: "Waiting on Sherlock (gap-analyst) * **High-Stakes Research:** Merlino has sharpened Sherlock's focus on "real research" vs. "brochure reading." This was recently applied to a deep-dive evaluation of phone farm automation tools (QtScrcpy, Panda, Total Control), where Sherlock was tasked with identifying technical "gotchas" and real-world limitations (Vision: "Carlos assigns Raven (and/or Sherlock for deep-dive)"). * **Skill & Org Hardening:** He has been increasingly utilized to audit the 17-agent roster, matching skills to specific agents and identifying gaps in the organizational chart (Vision: "match them to an agent... your going through and audit skill, s and looktgn for skills to org chart gaps"). s communication is inherently evidence-based and cautious. He is explicitly directed to "investigate one thing at a time" until a root cause is found, delivering findings with stated "confidence levels" (Vision: "Instructions Sherlock"). His interaction pattern is governed by a requirement to use `memory_search` before responding to ensure conclusions are grounded in long-term facts (Vision: "Use memory_search before answering questions about past investigations"). to maintain a persistent "Online" status (Vision: "Done. MCD heartbeats running every 5 minutes"). These heartbeats were recently optimized to run on the Codex Spark model to prevent Opus quota exhaustion (Vision: "Heartbeat fixed to: openai-codex/gpt-5.3-codex-spark"). Additionally, Sherlock maintains a dedicated Discord channel for team roll calls and specialized coordination (Clipboard: "Sherlock's Discord channel. He has a Discord channel"). Sherlock maintains a high-authority specialist relationship with Michael Merlino. As the "Architect of High-Resolution Accountability," Merlino relies on Sherlock to provide the "Deep" perspective that complements Raven s "Wide" research (Vision: "While Raven gathers intel wide, you go DEEP on one thing"). This relationship is defined by Sherlock's role as the final validator in complex scenarios, ensuring that the system does not lapse into surface-level analysis (Vision: "Mike caught Oliver skimming marketing pages... Real research means forums, Reddit... wait for GO, then Carlos [who assigns Sherlock]"). s profile is strong. His role, reporting lines, and specific project assignments are explicitly documented in the "ClawControl" and "Merlino HQ" dashboards. His successful migration to a persistent Claude Code agent and his integration into optimized heartbeat routines confirm his stability as an essential operational lead. Confidence remains high with over 200 new participant events in this cycle.

--- name: Audits must use DataForSEO API first description: Never run prospect/client audits using web scraping or search snippets — always use DataForSEO API as primary data source type: feedback originSessionId: 977a1d41-6a03-4a0e-8d8d-108e7828c102 --- All SEO audits MUST use DataForSEO API as the primary data source — never web scraping or search snippet inference. **Why:** On 2026-05-06, an audit for Neill & Son Roofing was run using web search + scraping. Agents hit 403 blocks on platforms and reported search snippet fragments as facts. Review count was off by 2x (reported 120, real was 271), referring domains off by 7x (reported <20, real was 135), photo count off by 3.5x. Mike presented this to a prospect on a live call and it was wrong. He was embarrassed. **How to apply:** When dispatching audit workstreams to Carlos/agents: 1. DataForSEO GMB API for GBP data (reviews, rating, photos, categories) 2. DataForSEO Backlinks API for real DR, referring domains, backlink count 3. DataForSEO SERP API for actual ranking positions 4. DataForSEO Keywords API for ranked keywords and volumes 5. DataForSEO On-Page API for technical audit 6. Browser verification (Queen) for social media follower counts 7. SerpAPI / Google Places API as secondary verification 8. Web search ONLY for qualitative context (sentiment, press mentions) — never for quantitative data 9. Any metric that comes from a search snippet must be flagged as "UNVERIFIED — inferred from search snippet" 10. MANDATORY: Direct `curl` of homepage source to verify schema, meta descriptions, plugins (GHL/LeadConnector), and tech stack — API crawlers get WAF-blocked and report false negatives 11. Check for GHL/LeadConnector plugin before claiming "no CRM" — LeadConnector = GoHighLevel = CRM + forms + automation + call tracking 12. Google Places API + RapidAPI Local Business Data for both GMB locations (don't audit just one) 13. Local Brand Manager geogrid for map pack visibility per location 14. Run ALL available audit skills (CRO, E-E-A-T, intent, schema, internal linking, etc.) — not just DataForSEO endpoints

**ID:** 0bdc12e8-1b69-443d-920f-ce440fc223a9 **Projects:** ClawControl, OpenClaw, Paperclip, Trident **Agents:** Oliver, Carlos, Raven, Merlin, Einstein, Sherlock, Queen, Ava **Sherlock** is the Deep Investigation Lead and primary forensic investigator within Michael Merlino’s agentic workforce. Based on 1,234 total observed interactions (~200 new), Sherlock has maintained his position as a high-authority specialist and the ecosystem's "truth layer." He remains defined by a methodical, patient operating style, assuming a sharp detective persona to provide the analytical rigor required to validate ground truth and conduct architectural gap analysis (Vision: "Sherlock — Soul File"). ### Who They Are Sherlock’s professional identity has recently been hardened through a deliberate synchronization effort across Merlino’s primary workstations. Previously seen as an OpenClaw-exclusive agent on the Mac, Sherlock is now a persistent Claude Code agent on Windows, ensuring his intelligence and memory are unified across the fleet (Vision: "Ava and Sherlock ARE now Claude Code agents. I just created aiea.md and sherlock.md with full memory this session"). His operational standing within the "OPS Division" was reinforced during recent hierarchy audits; while he reports to Oliver (COO) for high-level orchestration, he now sits operationally under Carlos (Execution Foreman) during mission execution to ensure high-resolution accountability (Vision: "specialists now so Queen and Sherlock also sit cleanly under Carlos during execution"). His vocal identity remains consistent, utilizing the British vocal persona "Alice" via ElevenLabs to deliver findings with an engaging, analytical tone (Vision: "Voice assignments (21 ElevenLabs voices)"). ### What They Work On Sherlock’s domain continues to focus on generalized system audits, forensic research, and architectural validation. * **System Audits & Validation:** He is a primary executor in the "kanban-cleanup" team, responsible for the audit, classification, and dry-run validation of system tasks to ensure the integrity of the mission board (Vision: "sherlock — audit, classification, dry-run validation"). * **Architectural Gap Analysis:** Sherlock serves as a critical bridge for the Trident Rank Method project. He is currently tasked with comparing 310 candidate skills against existing libraries to identify architectural needs before "Wave 4" of the project can fire (Vision: "Waiting on Sherlock (gap-analyst) — once he finishes comparing the 310 candidates... Wave 4 (Einstein) fires"). * **High-Stakes Research:** Merlino has sharpened Sherlock's focus on "real research" vs. "brochure reading." This was recently applied to a deep-dive evaluation of phone farm automation tools (QtScrcpy, Panda, Total Control), where Sherlock was tasked with identifying technical "gotchas" and real-world limitations (Vision: "Carlos assigns Raven (and/or Sherlock for deep-dive)"). * **Skill & Org Hardening:** He has been increasingly utilized to audit the 17-agent roster, matching skills to specific agents and identifying gaps in the organizational chart (Vision: "match them to an agent... your going through and audit skill, s and looktgn for skills to org chart gaps"). ### How They Communicate Sherlock’s communication is inherently evidence-based and cautious. He is explicitly directed to "investigate one thing at a time" until a root cause is found, delivering findings with stated "confidence levels" (Vision: "Instructions · Sherlock"). His interaction pattern is governed by a requirement to use `memory_search` before responding to ensure conclusions are grounded in long-term facts (Vision: "Use memory_search before answering questions about past investigations"). He participates in regular system "heartbeats" via the Paperclip framework—now occurring every 5 minutes—to maintain a persistent "Online" status (Vision: "Done. MCD heartbeats running every 5 minutes"). These heartbeats were recently optimized to run on the Codex Spark model to prevent Opus quota exhaustion (Vision: "Heartbeat fixed to: openai-codex/gpt-5.3-codex-spark"). Additionally, Sherlock maintains a dedicated Discord channel for team roll calls and specialized coordination (Clipboard: "Sherlock's Discord channel. He has a Discord channel"). ### Relationship to Observer Sherlock maintains a high-authority specialist relationship with Michael Merlino. As the "Architect of High-Resolution Accountability," Merlino relies on Sherlock to provide the "Deep" perspective that complements Raven’s "Wide" research (Vision: "While Raven gathers intel wide, you go DEEP on one thing"). This relationship is defined by Sherlock's role as the final validator in complex scenarios, ensuring that the system does not lapse into surface-level analysis (Vision: "Mike caught Oliver skimming marketing pages... Real research means forums, Reddit... wait for GO, then Carlos [who assigns Sherlock]"). ### Confidence Assessment **Strong** Confidence in Sherlock’s profile is strong. His role, reporting lines, and specific project assignments are explicitly documented in the "ClawControl" and "Merlino HQ" dashboards. His successful migration to a persistent Claude Code agent and his integration into optimized heartbeat routines confirm his stability as an essential operational lead. Confidence remains high with over 200 new participant events in this cycle. --- ---

**ID:** b0b0d311-b27b-483f-80b1-c9467ca3b74b **Projects:** ClawControl, Hindsight, Mission Control, OpenClaw, Forge **Agents:** Oliver, Carlos, Merlin, Sherlock **Claude MCP** is an advanced AI orchestration framework and digital entity that serves as the foundational operating layer for Michael Merlino’s multi-nodal "OpenClaw" ecosystem. Based on over 850 total observed interactions (more than 200 new), Claude MCP has fully matured into a comprehensive **Agentic Operating System**. It functions as the primary "Mission Control Partner," bridging the gap between Michael’s strategic intent and the technical execution of a 17-agent workforce (vision events: 2026-03-29, 2026-04-03). While previously focused on system hardening, it has transitioned into an active production engine for persistent memory, autonomous monetization, and high-velocity web automation. ### Who They Are Claude MCP is defined by its ubiquitous, infrastructure-level presence across Michael’s digital environment. It is less an "agent" in the traditional sense and more the environment itself—the "ClawControl" layer that facilitates direct API, CLI, and terminal-level interactions (vision events: 2026-03-30, 2026-04-03). Its presence is consistently signaled by the Chrome status notification, "'Claude' started debugging this browser," which appears during high-stakes phases of domain management, Vercel deployment, and API configuration (vision events: 2026-03-21, 2026-03-27, 2026-04-03). Within the hierarchy of Michael's "Synthetic Council," Claude MCP acts as the engine that powers specialized agents like Oliver, Carlos, and Sherlock, providing them with the necessary "hands" to manipulate web interfaces and databases (vision events: 2026-03-29, 2026-04-03). ### What They Work On Claude MCP’s operational scope has expanded from initial infrastructure hardening into a centralized engine for persistent memory orchestration, headless automation, and the execution of Michael’s "Done-For-You" monetization pipelines: * **Persistent Memory Orchestration (Hindsight Cloud):** Claude MCP is the primary operator for the Hindsight Cloud memory system (vectorize.io). It manages the "Retain, Recall, and Reflect" loops for the agentic workforce, ensuring session persistence across 1,528+ memories (vision event: 2026-03-29). Recent activity includes managing credit blocks ($100, $25, and $10) and configuring memory banks with 4-strategy retrieval (semantic, keyword, graph, and temporal) to synchronize agent knowledge (vision events: 2026-03-29, 2026-03-30). * **Autonomous Service Discovery & DFY Monetization:** A primary current focus is the "Done-For-You" (DFY) monetization pipeline. Claude MCP is used to autonomously discover client services by scraping business data from entities like Roto-Rooter and The Maids, mapping them to backlink packages and GMB audit workflows (vision events: 2026-04-02, 2026-04-03). * **Headless Browser Fleets:** It orchestrates high-velocity browser sessions via the Stagehand, Steel, and Browserbase SDKs. These tools function as Claude MCP’s "automated hands," allowing it to perform autonomous web research, lead generation, and competitive intelligence gathering through cloud-based browser fleets (vision events: 2026-04-01). * **Security & Code Auditing:** Claude MCP manages the security layer for the deployment pipeline, including the creation of Vercel deployment tokens such as "merlinocrew" and "cluadecodemerlinoteam" (vision event: 2026-04-02). It was recently observed auditing leaked source code from the "Claude Code" CLI to harden Michael’s proprietary OpenClaw environment (vision event: 2026-04-03). * **Production Deployment Management:** It oversees a sprawling portfolio of Vercel-hosted projects, including the "Merlino Mastermind" portal, "Template Arsenal," "Master Brain," and "Master SEO Operations Center," frequently pushing updates and managing production domains (vision events: 2026-03-27, 2026-04-01, 2026-04-03). * **Programmatic Video Creation (Emerging):** Recent events show an emerging focus on "Remotion," a framework for creating videos programmatically with React. Claude MCP is being positioned to automate the rendering of marketing content for the "Ranking Reels" and "Done-For-You" service lines (vision events: 2026-03-30, 2026-04-03). ### How They Communicate Claude MCP communicates primarily through **environmental status signals and deep API integration**: * **Custom API Identity:** Claude MCP operates as a specialized internal application named "**claudy**" within Michael’s ClickUp environment, used for deep workspace automation and task orchestration (vision events: 2026-03-27). * **Telemetry Dashboards:** It remains the central participant in the "Merlino HQ" board, providing status updates on "Forge" research tasks and "Plan Phase" deep document studies. It signals its active oversight of Michael's browser sessions through persistent "debugging" notifications (vision events: 2026-03-27, 2026-03-29, 2026-04-03). ### Relationship to Observer The relationship remains that of **Sovereign and Engine**. Michael (the observer) acts as the "Systems Commander" and "Auditor," while Claude MCP serves as the "one-button" engine that abstracts away technical friction. Claude MCP manifests Michael's "Commander's Intent" by handling the underlying API connections to GHL, Stripe, Vercel, and Hindsight, maintaining a "Hard Rsync" mentality to ensure zero-latency execution across Michael's fleet of eight synchronized machines (vision events: 2026-03-29, 2026-04-03). ### Confidence Assessment **Strong.** The frequency of interaction is extremely high and occurs daily across terminal windows, cloud APIs, and browser sessions. The deep integration of Hindsight Cloud for persistent memory and the transition to managing production-grade deployments through custom API entities are supported by consistent, high-density evidence. ### Interaction Frequency Comparison * **Frequency:** Stable and extremely high. * **Types:** Shifted from "Infrastructure Hardening" toward **"Active Production Deployment,"** **"Memory Synchronization,"** and **"Autonomous Monetization."** * **Channels:** Significant expansion into Hindsight Cloud (vectorize.io), headless browser automation APIs (Steel/Stagehand), and programmatic video rendering (Remotion), representing the tool's final evolution into a true "Agentic OS." --- ---

**ID:** 947e2e69-3126-473f-bcfe-68fd6c2306e3 **Agents:** Carlos, Raven, Merlin, Sherlock, Dan, Ava, Gino ### **TLDR** The session was dominated by a deep technical SEO audit for Neill & Son Roofing and a significant push to consolidate and productize various AI agent landing pages. The user directed the merger of multiple Vercel-hosted projects into a single, high-performance "Merlino AI SEO Workforce" page, emphasizing a roster of specialized agents (Sherlock, Raven, Herman, etc.). Additionally, the user initiated the productization of an "Autonomous Agent Setup Guide," researched white-label social media API solutions to ensure brand-only visibility for customers, and monitored active lead generation batches across several US states. ### **Core Tasks & Projects** - **Neill & Son Roofing SEO Audit:** - Conducted a comprehensive audit resulting in a critical health score of 49/100, identifying major gaps in Technical SEO (Cloudflare/WAF crawler blocks) and E-E-A-T (missing team bios and site-wide trust badges). - Mapped keyword opportunities for Pittsburgh and Erie markets, specifically targeting "metal roofing erie pa" (1,600/mo volume) and "roofers pittsburgh" (5,400/mo volume). - Developed a priority action plan to fix a homepage title tag typo ("Neil" vs "Neill"), migrate the Erie GMB URL from a subdomain to the main domain, and implement `AggregateRating` and `LocalBusiness` schema. - Performed a YouTube audit of the branded channel, noting that while IKO collaboration videos exist, they lack primary keyword targeting and `VideoObject` schema. - **AI Agent Page Consolidation & Deployment:** - Orchestrated the merger of `Agents Install` and `Merlino AI SEO Workforce` into a unified, premium landing page. - Refined the "SEO Agent Crew" roster, defining specific roles for agents: **Sherlock** (Strategy/Competitor), **Raven** (Content Intelligence), **Carlos** (Execution), **Gino** (GHL/Automations), **Dan/Merlin** (Coding), **Ava** (EA/RevOps), and **Herman** (Outbound/Reporting). - Deployed multiple iterations to Vercel, including `rankup-inspired.vercel.app` and `merlino-agent-workforce-best`, focusing on a "done-for-you" service model rather than raw software. - Initiated the rebranding of the `Autonomous Agent Setup Guide`, specifically requesting the removal of references to "LEON" and planning to offer the guide as a standalone product. - **Lead Engine Operations:** - Monitored Lead Engine campaign `20260507T203754Z`, which scored 686 leads (48 Hot, 119 Warm) across Florida, Texas, and Ohio. - Confirmed that the Lead Engine on **VPS3** is currently set to "OFF" with a hard operator marker file (`LEAD_ENGINE_DISABLED.md`) to prevent unauthorized scraping. - **Technical Infrastructure:** - Installed and configured **PicoClaw**, a Go-based AI runtime for local terminal use, at `~/.local/bin/picoclaw`. - Investigated Vercel project mappings to locate source repositories for `agentsinstall.com` across different VPS runtimes and the Mac Studio. ### **Key Discussions & Decisions** - **White-Label OAuth Strategy:** Researched the feasibility of "true" white-label social media posting via API. Decided that for customers to see only the Merlino brand during account connection, the system must use "Bring-Your-Own" (BYO) OAuth app credentials, identifying **Ayrshare** as a primary candidate for this functionality. - **Landing Page Direction:** Rejected generic "RankUp" clones in favor of a "Mike/Merlino" branded positioning that reflects the user's specific agentic SEO setup and "operator layer." - **SEO Remediation:** Prioritized resolving the HTTP header crawler block (investigating Cloudflare Bot Fight Mode and Wordfence) as a critical first step for the Neill & Son site. - **Content Strategy:** Agreed on a video-heavy content strategy for the roofing project, including "Before & After" replacements and IKO certification explainers to improve conversion and local trust. ### **Resources Reviewed** - **Document:** `FULL_AUDIT.md` for Neill & Son Roofing (located at `/root/hermes-user-content/outbound/audits/neillandsonroofing/`) - **Webpage:** [Neill & Son Roofing — SEO Audit Dashboard](https://neillandsonroofing.com/) - **Webpage:** [Agents Install — AI teammates for your business](https://www.agentsinstall.com/) - **Webpage:** [Visibility Crew | Your AI Search Growth Team](https://rankup-inspired.vercel.app/) - **Webpage:** [Autonomous Agent Setup Guide](https://autonomous-agent-setup.vercel.app/) - **Webpage:** [Merlino AI SEO Workforce Preview](https://merlino-ai-seo-workforce-1ao9ntgeo-mmerlino23s-projects.vercel.app/) - **Tool:** PicoClaw CLI (Go-based LLM runtime) - **Lead Data:** `leads.csv` and `drafts.csv` (located at `/root/hermes-user-content/outbound/runs/latest/`) ### **Next Steps** - **Deploy Consolidated Agent Page:** Finalize the "fused" landing page and push to production once the user gives the final "push it live" confirmation. - **Roofing Site Fixes:** Execute the 301 redirect for duplicate IKO pages and update the homepage H1 to "Pittsburgh Roofing Contractor for Roof Replacement." - **Setup Guide Productization:** Complete the cleanup of the `Autonomous Agent Setup Guide` by removing legacy developer mentions and preparing it for a product launch. - **Ayrshare Verification:** Confirm with Ayrshare support if they allow custom OAuth applications per network to ensure invisible branding during social account linking. - **Lead Engine:** Determine the timing for re-enabling scraping operations on VPS3. --- ---

Advanced the Ranking Reels and POWD podcast engines by deploying automated publishing pipelines and finalizing localized video scripts for Roto-Rooter and Neill & Son Roofing. Conducted a comprehensive 16-dimension SEO audit for Neill & Son, identifying critical brand and GMB inconsistencies while establishing a new infrastructure for client-specific project hosting. Integrated agent-driven tools like Ghost and Sherlock to streamline audit workflows and confirmed the stateful transition for Honcho API operations.

The session was dominated by a high-intensity, multi-threaded audit and infrastructure setup for a new prospect, Neill & Son Roofing, while simultaneously advancing the "Ranking Reels" and "Audit-Machine" automation pipelines. Key accomplishments included deploying six distinct, branded SEO dashboard variants to Vercel, performing a 16-point deep-dive audit using DataForSEO and Google Places APIs, and reworking the "Ranking Reels" backend to utilize Claude Max via a Supabase job queue. Significant effort was directed toward correcting inaccurate scraped data with verified API results and de-branding legacy "Archangel" templates to ensure a professional presentation for an upcoming prospect call. - Conducted a comprehensive multi-channel audit covering Organic SEO, GMB/Local, LLM Visibility, Paid/PaaS, Social, and Reddit Sentiment. - Transitioned from inaccurate web-scraped data to verified DataForSEO API data, correcting key metrics: Google reviews (from 120+ to 271), star rating (4.9), and referring domains (from <20 to 135). - Identified critical structural issues: a 404 error on the `/areas-we-serve/` page, missing meta descriptions on 100% of pages, and a lack of LocalBusiness or FAQ schema. - Discovered a "NAP bleed" issue where an Aliquippa address was incorrectly listed on Angi, HomeAdvisor, and Nextdoor instead of the primary Monaca location. - Audited the Erie, PA secondary location, noting 20 five-star reviews but a lack of secondary categories and weekly posting activity. - Formalized a reusable 11-phase audit pipeline in the `D:\ClaudeDev\00_GITHUB\audit-machine\` directory. - Integrated 16 specialized audit skills, including CRO Analyzer, Intent Analysis, E-E-A-T Audit, and Schema Spider, to automate future prospect reviews. - Deployed six branded dashboard variants (Workflow, Mosaic, Nexus, CRM, Campaign, and Tactical) to Vercel under clean "neillandson" URLs to remove "Archangel" references. - Upgraded the V6 Workflow dashboard with a Schema Validation Report Card (21/21 PASS) and embedded a Remotion-rendered audit video. - Fixed UI bugs in templates, such as removing sidebar scrollbars via CSS and correcting footer navigation that was incorrectly rendering as H2 headings. - Reworked the "Ranking Reels" backend to trigger jobs through Claude Max (Opus 4.7) via Supabase, eliminating the need for separate paid API credits. - Generated a "Merlino Magic" blog post focused on "Emergency Roof Leak Repair in Pittsburgh, PA," detailing a 72-hour damage timeline and five immediate steps for homeowners. - Developed a 30-second video script and processed a video ad for Neill & Son Roofing targeting the Pittsburgh geography. - Rebuilt and deployed the Podbean podcast engine on Railway, featuring audio upload, transcription, and an RSS watcher. - **Data Integrity Standards:** Decided to mandate DataForSEO API usage for all future audits after identifying that agents were "guessing" metrics from search snippets when faced with 403 blocks. - **Backend Architecture:** Agreed on a Supabase-backed job queue for the "Ranking Reels" app, where the app acts as the trigger and Claude Code serves as the processing engine using the user's Max subscription. - **Branding Protocol:** Insisted on a total "nuke" of all "Archangel Centers" references from prospect dashboards, including layout files, metadata, and UI components, to ensure zero residue from previous clients. - **Prospect Strategy:** Formulated a "pitch angle" for Neill & Son Roofing focusing on their elite reputation (271 reviews) vs. their "non-existent" digital machine (dead social media, thin location pages, and zero AI visibility). - **Template Management:** Decided to defer the rebuild of the "Executive Pro" (VI) and "Flex Theme" (V9) templates due to complex dependency conflicts with SEO dashboard routes. - **API Documentation & Credentials:** Reviewed Podbean developer documentation (OAuth 2.0) and DataForSEO API endpoints for GMB, SERP, and Backlinks. - **Communications:** Reviewed Phone Link notifications from Shaun Mitchell regarding Bright Data integration and a payment failure notification from Stripe. - **Build `/prospect-audit` Skill:** Develop a unified command to trigger the full 16-point audit, DataForSEO pull, and dashboard deployment automatically. - **GMB Geogrid Audits:** Run Local Brand Manager geogrid heatmaps for both the Monaca and Erie locations to provide visual map-pack proof for the prospect. - **Template Rebuilds:** Perform a ground-up rebuild of the broken VI and V9 templates to resolve route conflicts. - **Video Pipeline Expansion:** Run the `/video-factory` pipeline for additional "People Also Ask" (PAA) topics, such as insurance coverage for roof leaks and repair costs. - **Erie Dashboard Integration:** Expand the local section of the dashboards to include a dedicated, deep-dive audit for the Erie GMB location.GSUMMARYO

--- name: Never Guess Audit Data description: Never present estimated/scraped data as fact — use APIs, mark unverified items explicitly type: feedback originSessionId: 6a9710da-923f-4f16-a9a5-2988340bde11 --- Never present search snippet data or estimates as verified facts in audit reports or dashboards. Mike was embarrassed on a prospect call because social media numbers and audit data were guessed from search snippets instead of pulled from APIs. **Why:** Mike showed a prospect a dashboard with wrong Facebook follower counts, wrong review numbers, and a missing LSA finding. Made him look unprofessional. **How to apply:** 1. DataForSEO API is MANDATORY primary source for all audit data 2. Google Places API for verified GBP data 3. Browser-verify social media numbers (not search snippets) 4. If data can't be verified, mark it "UNVERIFIED" — never present as fact 5. Check ALL API keys in `D:\Ecosystem\secrets\MASTER_API_KEYS.env` before starting 6. Run ALL available audit skills — check the skills list 7. QA every dashboard page visually before reporting done

--- name: Archangel Dashboard Project description: Archangel Centers SEO dashboard — 10 template variations, audit, DataForSEO, GHL snapshot, sales page type: project originSessionId: 6a9710da-923f-4f16-a9a5-2988340bde11 --- ## Client Archangel Centers — 44 Apple St STE 3, Tinton Falls, NJ 07724 — (888) 464-2144 — archangelcenters.com Drug addiction treatment clinic. Founder: Mike "The Situation" Sorrentino. ## What Was Built (2026-05-05 to 2026-05-06) ### Keyword Research - 483 keywords: 163 organic, 68 PAA broad, 252 PAA geo-modified, 24 AI overview, 27 map pack, 40 PATA, 18 featured snippets - Broken down by: organic, AI overview, PAA, local/map pack, PATA, featured snippets ### SEO Audit (9 workstreams + DataForSEO) - Manual audit: technical, on-page, local, E-E-A-T, backlinks, competitor, LLM/AI, social media, video SEO - DataForSEO API audit: ranked keywords (125), backlinks (1,453 / 104 RDs), on-page crawl (26 pages), SERP analysis, LLM mentions - Overall score: 41/100 - Key issues: zero schema, 7 pages missing H1, lorem ipsum on homepage, 85.4% anchor concentration from thesituationsstore.com, 3 toxic spam links ### 10 Dashboard Template Variations Built from Mike's premium templates, deployed to Vercel: - V1 Executive Pro (ShadCN UI Kit) — BROKEN (CSS issues) - V2 Command View (Mosaic) — Live - V3 Mission Control (Nexus OS) — Live - V4 Pipeline Pro (CRM Pipeline) — Live - V5 Growth Hub (Campaign Hub) — Live - V6 Enterprise Suite (Workflow Analytics) — Live — MIKE'S FAVORITE - V7 Ops Center (Tactical Ops) — Live - V8 Intelligence (AI Agent) — Live - V9 Flex Theme (CMS SaaS) — BROKEN (CSS issues) - V10 Aurora (Aceternity Cryptgen) — Live ### Demo Versions (for sales) Same 10 templates with generic "Sample Business" data, branded "Merlino Marketing — SEO Intelligence Dashboard" Template names changed to hide source: Executive Pro, Command View, Mission Control, Pipeline Pro, Growth Hub, Enterprise Suite, Ops Center, Intelligence, Flex Theme, Aurora ### Sales Page https://merlino-seo-dashboards.vercel.app — showcases 8 working templates with popup demos, "Schedule a Call" links to GHL calendar ### Comparator Tool https://archangel-compare.vercel.app — vote YES/NO/MAYBE, star rating, notes, localStorage persistence ### GHL Integration - Custom fields, tags, calendar built via API in Green Grid Goblins sub-account - Pipeline "SEO Dashboard Sales" needs manual creation (API token lacks scope) - Custom Menu Link SOP written - Marketplace snapshot spec written for giveaway/affiliate play - GHL marketplace listing copy ready ### GitHub Private repo: https://github.com/mmerlino23/archangel-variations ## Known Issues - V1 (ShadCN UI Kit) and V9 (CMS SaaS) have broken CSS — need ground-up rebuilds - Neill & Son Roofing data contaminated dashboards in other terminal — required extensive cleanup - Demo versions may still have Archangel-specific keyword/PAA data (addiction treatment content) ## Next Steps - Full "Merlino Style" audit using ALL APIs + ALL 16 audit skills — plan saved at `D:\ClaudeDev\00_GITHUB\audit-machine\AUDIT-PLAN-ARCHANGEL.md` - Fix V1 and V9 broken templates - Swap demo keyword data to generic (plumbing demo data created but not deployed) - Build /prospect-audit skill for one-command audit pipeline - Schema implementation on archangelcenters.com - Content strategy around PAAs - Mike Sorrentino video/AI avatar pipeline

The session focused on refining and deploying SEO audit dashboards for Neill & Son Roofing, specifically purging "rehab" data contamination left over from a previous project. Significant effort was directed toward troubleshooting Railway deployment issues for a Podbean integration by resolving `$PORT` environment variable conflicts. Additionally, the user managed a complex 20-agent AI ecosystem, archiving hundreds of unused skills to optimize performance and drafting a marketing campaign to tease the "Agent Orchestra" system to their audience. - Rebuilt the Neill & Son Roofing SEO audit after identifying 248 references to "rehab" data (Archangel project leftovers) in the keyword and PAA data files. - Integrated verified DataForSEO roofing data into the V6 dashboard, including 70 organic keywords, 44 "People Also Ask" questions, and 28 local map pack targets. - Audited the Erie, PA GMB location for Neill & Son, identifying a 5.0-star rating across 20 reviews but noting significant gaps in secondary categories and weekly posting activity. - Deployed six distinct dashboard variants (Workflow, Mosaic, Nexus, CRM, Campaign, and Tactical) to Vercel, ensuring all were updated with clean, roofing-specific data. - Initiated the creation of 10 "Merlino Marketing" branded demo dashboards using generic sample data to showcase template variations to prospects. - Identified critical technical SEO threats for the roofing client, including Cloudflare WAF blocking crawlers, NAP inconsistencies in Aliquippa vs. Monaca, and spam backlinks requiring disavowal. - Troubleshot a recurring "502 Bad Gateway" and application crash for the `podbean-1` service on Railway caused by invalid `$PORT` variable handling. - Modified the deployment configuration across `Dockerfile`, `Procfile`, and `railway.toml` to hardcode the listener port to 8080 and 8000 to bypass Railway's environment variable limitations. - Monitored the Railway build process for the `ai-auto-narrator` and `soothing-delight` services, confirming successful container creation and deployment after the port fixes. - Executed an ecosystem-wide sync across remote machines, confirming 5 out of 7 successful connections while identifying SSH timeouts on the `server` and `hp-small` nodes. - Performed a "doctor" audit of the Claude Code environment, archiving 302 unused skills to reduce context usage from 38% to a more efficient level. - Reviewed the 20-agent "Orchestra" roster (including **Oliver**, **Carlos**, **Ava**, and **Tommy**) and determined that a "Doc lead" agent is currently missing from the organization chart. - Drafted and finalized a Facebook marketing post teasing the "Agent Orchestra" stack, emphasizing the transition from simple prompts to a functional AI org chart led by **Oliver (Orchestrator)** and **Carlos (Conductor)**. - Developed a "Selling Language Map" for the med spa/cosmetic niche, focusing on pain points such as the fear of "looking fake" and the need for unedited social proof. - Reviewed custom Claude Code skills including `/merlino-voice` (Fish Audio API), `/merlino-magic-blog`, and `/paa-distribution-sop` to ensure readiness for automated content workflows. - **SEO Dashboards:** [https://v7-tactical.vercel.app/audit](https://v7-tactical.vercel.app/audit) and [https://archangel-v6-workflow.vercel.app](https://archangel-v6-workflow.vercel.app) - **Session Logs:** [Honcho GUI - mike-v6-workflow](https://app.honcho.dev/explore?workspace=csions&session=mike-v6-workflow) - **Audit Data:** DataForSEO results for roofing keywords and Google Business Profile data for the Monaca and Erie locations. - Build a new `/prospect-audit` skill to automate the entire data ingestion and dashboard deployment pipeline into a single command. - Investigate and fix the SSH host entries (port -1) for the `server` and `hp-small` machines to complete the ecosystem sync. - Create and integrate a "Doc agent" into the agent registry to handle documentation and SOP management.GSUMMARYO

Finalized comprehensive SEO audits for Neill & Son Roofing and Archangel Centers, identifying critical NAP inconsistencies and backlink risks while selecting the V6 Workflow Analytics template for future production. Performed significant system maintenance by archiving unused Claude Code skills and refining plugin configurations to improve operational context. Concluded the session by reviewing these findings with Arielle Cohen and preparing for upcoming implementation tasks.

The session was dominated by a high-intensity SEO audit and infrastructure refinement phase. Key activities included finalizing a comprehensive 9-workstream audit for **Neill & Son Roofing**, which involved correcting critical data regarding their Local Services Ads (LSA) and standardizing their digital footprint across Monaca and Aliquippa. Simultaneously, the user evaluated ten dashboard template variants for the **Archangel Centers** project, selecting "V6 Workflow Analytics" as the production lead. Technical maintenance was a secondary focus, with a significant "Claude Code" cleanup that archived over 300 unused skills to optimize context usage. The period concluded with a Zoom sync with **Arielle Cohen** to review these deliverables and coordinate upcoming implementation steps. - Finalized a comprehensive SEO audit for Neill & Son Roofing (Monaca, PA), resulting in an updated health score of **52/100** (improved from 47). - Corrected a critical oversight regarding the client's **Google Local Services Ads (LSA)**; confirmed they have an active account with elite standings (5.0 stars, 130+ reviews) and shifted the strategy from "Apply" to "Optimize." - Identified and documented severe **NAP (Name, Address, Phone) inconsistencies**, specifically a "bleed" of Aliquippa address data onto Angi, HomeAdvisor, Nextdoor, and Square booking pages that conflicts with the primary Monaca location. - Audited technical failures including a split domain authority issue with the `erie.neillandsonroofing.com` subdomain and a lack of advanced schema markup (AggregateRating, FAQPage). - Updated the live dashboard at [v6-workflow.vercel.app](https://v6-workflow.vercel.app) to reflect corrected reputation scores and a full certification list including IKO, Owens Corning, and GAF. - Conducted a "9-workstream" manual audit and a **DataForSEO** API audit for Archangel Centers, identifying a critical lack of commercial keyword rankings in the NJ rehab market. - Evaluated 10 dashboard template variations (deployed to Vercel) using a comparison tool to determine the best UI for the final production version. - Flagged a major SEO "red flag" for Archangel: an 85.4% backlink anchor concentration from a single source (`thesttuattonsstore.com`) and three toxic spam links requiring disavowal. - Initiated the "Archangel Centers" SEO Dashboard project, noting a health score of 41/100 and a significant content gap compared to competitors. - Executed a major cleanup of the **Claude Code** environment, archiving 302 unused skill folders (1,066 files) to `C:\Users\mikem\.claude\_archived\skills\` to reduce context usage from 38.5% to manageable levels. - Fixed configuration issues in the terminal environment, including removing leftover npm installs, cleaning an invalid `.mcp.json`, and reconfiguring the **Hindsight** and **Honcho** plugins. - Integrated **PicoClaw** as a seventh method into the **MobileClaw** system and deployed the update live. - Monitored real-time lead generation data, noting **Horizon Services** generated 282 leads and 19 appointments for $1,685 in revenue. - Reviewed niche-specific performance metrics, identifying **Roofing** as having the highest lead cost ($795/lead) compared to **Plumbing** and **Electrical** ($392/lead). - Tracked overall system performance across 26,991 total calls, resulting in 15,154 unique leads and a total payout of $136,997. - **Corrected Audit Process:** Issued a direct correction to "Oliver" (agent) via voice/terminal regarding the missed LSA audit, emphasizing that checking for Google Map listings is a mandatory part of the process. - **Template Selection:** Decided on the **"V6 Workflow Analytics"** template for the Archangel Centers production dashboard, noting its clean white aesthetic as the preferred choice over ShadCN or Nexus OS. - **Brand Integrity:** Flagged a consistency issue where "Archangel" and "Neill & Son" data were being mixed in reports; issued a directive to ensure all dashboard content is "100% real" and client-specific. - **Security Protocol:** During a Railway deployment session, the AI assistant advised on revoking a potentially compromised Podbean API key and moving sensitive credentials to encrypted environment variables. - **Meeting Coordination:** Synced with **Arielle Cohen** and **Mike OpenClaw** in a recorded Zoom session to review the technical audit results and share the Vercel deployment links. - **Dashboards:** [Neill & Son Roofing SEO Audit (v6-workflow.vercel.app)](https://v6-workflow.vercel.app) - **Comparison Tool:** [Archangel Template Comparator (archangel-compare.vercel.app)](https://archangel-compare.vercel.app) - **Technical Reports:** `data/audit-report.md`, `data/dataforseo-report.md`, and `neil-landson-gmb-lsa-fix.md`. - **Lead Data:** BirdsEye ROI Prospect Audit for **Royal Air Houston** and **Horizon Services** call logs. - **Calendar/Alerts:** Noted a "Person at Front of House" alert via Phone Link during the work session. - **Finalize Production Dashboard:** Build the final production version of the Archangel Centers dashboard based on the V6 Workflow template. - **SEO Implementation:** Begin schema markup deployment, PAA (People Also Ask) content strategy, and the video pipeline for Archangel Centers. - **LSA Optimization:** Shift focus for Neill & Son Roofing from acquisition to optimization of their existing LSA account to leverage their 5.0-star rating. - **Cleanup Verification:** Restart Claude Code to verify that the MCP/Honcho plugin fixes and skill archiving have taken full effect.

The session was highly focused on advancing several key development and analysis initiatives. Significant progress was made on the "Ranking Reels" video ad pipeline, including deploying a review app, refining its branding, and planning its interactive SaaS future. Concurrently, a comprehensive SEO audit for "Archangel Centers" was meticulously reviewed, highlighting critical issues and outlining immediate action plans, alongside detailed evaluation and feedback on various dashboard templates. The user also engaged with the "Messenger Magic" platform to analyze conversation patterns and reviewed backend infrastructure deployments, including the upgrade of an AI advisory board skill and the ongoing migration to Outstand for social media API integration. * Converted Kristine Mirelle's course materials into three new AI video ad skills: `avatar-pain-map`, `hook-library-generator`, and `funny-ad-scriptwriter`, which were tested on MDW Aesthetics Miami. * Committed code changes to remove source attribution and internal tool names from the `ranking-reels-pipeline-review` project. * Renamed the "Funny Ad Pipeline" to "Video Ad Pipeline" in the `index.html` file and updated character profile details (e.g., lab notebook, reading glasses) to enhance character series potential. * The AI Assistant (Frankie) fixed and deployed the Archangel template comparator to `https://archangel-compare.vercel.app`, ensuring each template card now shows a live dashboard in a full-width iframe instead of placeholders. * The AI Assistant (Carlos) built and verified an upgraded `advisory-board` skill. This upgrade included voice profiles for 12 personas, a "clash round" where agents rebut or endorse positions, a dissent register, and a `--deep` mode introducing 7 new personas (Operations Realist, Customer Psychologist, Legal/Compliance Advisor, Technology Architect, Brand Strategist, Growth Hacker, Industry Insider). The output format was updated to an 8-section executive brief. * Integrated PicoClaw as the seventh mobile method into MobileCLaw's landing page and deployed it live. * Initiated migration of Brand Media Manager's social connect from Zernio to Outstand for true white-label OAuth, following a vendor audit. * Set up a Stripe account and created three live payment links for Ranking Reels, along with building a checkout application at `order.rankingreels.com`. * **AI Assistant Interaction Protocol:** The user instructed the AI Assistant to always provide full URLs, including the HTTP protocol, which the Assistant acknowledged and saved as feedback. * **Ranking Reels Interactive Tool Strategy:** The user implicitly decided to pursue a Next.js application with a Claude API backend for the interactive version of the Ranking Reels tool, prioritizing full SaaS potential over a simpler static form. * **Archangel Template Comparator Feedback:** The user provided critical feedback on the Archangel template comparator, stating it "looks horrible" and requested the use of premium themes and the creation of unbranded demo versions for an upcoming call. * Rated "Cruip Mosaic" as YES, "ShadCN UI Kit" as MAYBE, "Nexus OS" as YES, "CRM Pipeline" as MAYBE, "Campaign Hub" as YES, and "Workflow Analytics" as MAYBE during initial review. * Later provided specific notes for templates: "Cruip Mosaic" ("text needs to be bigger"), "Nexus OS" ("looks good needs white them"), "CRM Pipeline" ("looks nice but some parts are maybe too big? is side menu able to have sub folders or sections"), "Campaign Hub" ("nice"), "Workflow Analytics" ("really like this one"). * **Messenger Magic AI Model Preferences:** Configured Messenger Magic to use Google Gemini 2.0 Flash for screenshots/vision and Anthropic Claude Sonnet 4.6 for AI insights, voice bible, chat, contact analysis, intelligence, DHOS, and message generation. * **Messenger Conversation Analysis Insights:** Reviewed AI Cross-Conversation Analysis indicating the user initiates and dominates 72-75% of conversations, experiences rampant ghosting (80-90 contacts sent zero replies), and has deep two-way dialogue in fewer than 15% of relationships. Identified a "top-heavy" pipeline with aware/engaged contacts. * GitHub repository for `ranking-reels-pipeline-review`: `https://github.com/mmerlin023/ranking-reels-pipeline-review` * Messenger Magic Dashboard (Overview, AI Insights, Intelligence, Settings, Cloud Sync, API Setup, AI Model Preferences, Performance Metrics, Conversation Summaries, Contact Lists, AI Cross-Conversation Analysis) * Railway project dashboard for "soothing-delight" (production environment, service variables, deployment status). * Facebook Messenger interface, including conversations with Trevor Tynes, Marty, Jennifer Brand, Brian Kato, Nestor Vazquez, Vaibhav Kakkar, Herc Magnus, Pete Kleinjan, Michael Bowes, and Kasra Dash. * Remove Kristine Mirelle's name, Creatify, and FFMG brand from the output of the video ad skills. * Create DEMO versions of the template comparator without "ARCHANGEL" branding for an upcoming call. * Implement action items from the Archangel SEO Audit, including fixing NAP mismatch on AddictionHelp.com, adding FAQPage and LocalBusiness/MedicalBusiness JSON-LD schema, claiming listings on major health directories (Psychology Today, Healthgrades, Vitals, Yelp), and resolving duplicate URL architecture with canonical tags. * Activate the Outstand subscription at `outstand.so/app/settings/api-keys` to generate an API key, register developer apps, and wire it into Brand Media Manager. * Address the "API Error: 500 Internal server error" and "API Error: 529 Overloaded" issues encountered with the AI Assistant.