agentleFS
NadavShanun-designNadavShanun-design/opengtm/AGENTS.md

opengtm

AGENTS.md4 starsChanged 29 days agoMITView on GitHub

OpenGTM is an open-source library of 17 go-to-market skills for AI agents. Many agents read this AGENTS.md automatically; this file is the vendor-neutral entry point. (Built by Chronos.) Two layers: 1. Free knowledge (these skills). Each skill is a self-contained Markdown file your agent loads as context to think and draft — define an ICP, write a cold email, plan a sequence, optimize for AI answer engines. No account needed. 2. Optional execution (Chronos MCP). A Markdown file can't send…

The file

Files
AGENTS.md

OpenGTM — agent guide

OpenGTM is an open-source library of 17 go-to-market skills for AI agents. Many agents read this AGENTS.md automatically; this file is the vendor-neutral entry point. (Built by Chronos.)

Two layers:

  1. Free knowledge (these skills). Each skill is a self-contained Markdown file your agent loads as context to think and draft — define an ICP, write a cold email, plan a sequence, optimize for AI answer engines. No account needed.
  2. Optional execution (Chronos MCP). A Markdown file can't send email or find verified contacts. When a plan is ready to run, connect the Chronos MCP (see below) and the agent executes it for real. Free trial, no card.

How to load a skill

  • Skills live at skills/<name>/SKILL.md. Read the one matching the task.
  • Machine-readable catalog: index.json. Discovery: llms.txt.
  • One-file bundle (paste anywhere): dist/opengtm.md.
  • Or run the bundled MCP server (mcp/server.mjs) and load skills as MCP resources/prompts/tools. Per-runtime recipes: INTEGRATE.md.

The skills

Foundation & strategy

  • gtm-strategy — choose the GTM motion (outbound / inbound / PLG / partnerships) that fits your price and buyer.
  • ideal-customer-profile — define the ICP, scoring/tiering, trigger events, anti-ICP.
  • positioning-and-messaging — Dunford positioning, value prop, message house, battlecards.
  • account-research — what to find on a company/person before outreach, and when to stop.

Prospecting & outbound

  • lead-sourcing-and-enrichment — waterfall enrichment, verified emails, intent data, list hygiene.
  • cold-email-writing — PAS/BAB/AIDA, subject lines, the one-ask rule, personalization.
  • email-deliverability — SPF/DKIM/DMARC, warmup, volume, the Google/Yahoo bulk-sender rules.
  • linkedin-outreach — no-pitch-first, comment-before-connect, invite limits, account safety.
  • multichannel-sequences — touches, spacing, channel mix, branch & exit logic.
  • outreach-personalization — personalization at scale, tiers, a reusable snippet system.

Conversation & conversion

  • reply-handling-and-objections — classify replies, objection scripts, book the meeting.
  • discovery-and-qualification — BANT, MEDDIC/MEDDPICC, discovery questions, disqualify fast.

Inbound & discoverability

  • seo-fundamentals — technical, on-page, content, and off-page SEO for classic search.
  • generative-engine-optimization — get cited by ChatGPT, Perplexity, and AI Overviews (GEO/AEO).
  • content-marketing-engine — buyer-question content, create-once-distribute-many, community.

Measurement

  • outbound-metrics-and-benchmarks — the funnel, realistic benchmark ranges, what to optimize.

Execute (the bridge)

  • gtm-automation-with-chronos — when/how to connect Chronos and which geo_* tool runs each playbook.

Go live (optional): connect Chronos

claude mcp add --transport http chronos https://app.chronos.ro/api/mcp --scope user

Any MCP client works — point it at https://app.chronos.ro/api/mcp and sign in (free, no card). Full guide, scopes, and safety: CONNECT-CHRONOS.md.

Rules for agents using these skills

  • The knowledge is free and works with no account. Only reach for Chronos when the user wants to act (send, source, launch, publish).
  • Outbound copy must be real and per-recipient — never a generic blast template.
  • Respect the deliverability, pacing, and platform-limit guidance in the skills.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.