Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/da-troll/planhat-mcp/agents-mdgit clone --depth 1 https://github.com/da-troll/planhat-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/da-troll/planhat-mcp/agents-md)<a href="https://agentmods.dev/instructions/da-troll/planhat-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/da-troll/planhat-mcp/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01813 | $0.01813 |
| Opus 5 | $0.00907 | $0.00907 |
| Sonnet 5 | $0.00363 | $0.00363 |
| Haiku 4.5 | $0.00181 | $0.00181 |
Grade A, and why
planhat-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md: handbook for AI coding agents (and curious humans)
This file is the single source of agent guidance for this repo. CLAUDE.md is a
symlink to it; edit this file only.
What this repo is
A small TypeScript MCP server exposing the Planhat CRM REST API
(https://api.planhat.com) as 60 tools: 12 resource families x 5 verbs
(list/get/create/update/delete). Built on the official @modelcontextprotocol/sdk,
using Node's built-in fetch, with a bearer token read from the environment (or a
local .env for manual installs).
esbuild compiles src/ into one dependency-free dist/server.js. The bundle
manifest is server.type: "node", so Claude Desktop runs it with its own built-in
Node runtime: users install nothing (no Python, no uv, no runtime). Node.js 18+ is
required only for manual installs from a checkout.
Commands
npm install # install dependencies
npm test # offline test suite (node:test); all tools, fetch mocked
npm run typecheck # tsc --noEmit, strict
npm run build # esbuild -> dist/server.js
npm start # run the built server over stdio
A read-only live smoke check (needs a real token in .env; GETs only, mutates nothing):
node -e '
const t = process.env.PLANHAT_TOKEN;
const paths = ["/companies","/endusers","/opportunities","/conversations","/users","/assets","/issues","/tickets","/tasks","/licenses","/invoices"];
for (const p of paths) {
const r = await fetch("https://api.planhat.com"+p+"?limit=1", { headers: { Authorization: "Bearer "+t } });
console.log(r.status, p);
}'
Hard rules
- Never commit
.envorplanhat-mcp-notes.md. Both are gitignored; they contain live credentials. If you ever see a secret in a tracked file, stop and flag it. - Never fire create/update/delete calls against the live API in tests, smoke
checks, or "just to verify". This is a production Planhat tenant. Read-only GETs
with
limit=1are the ceiling for live verification. - Tests must stay fully offline: tool tests drive handlers with a recorder Http, and
the HTTP tests mock the global
fetch. - No emojis and no em dashes in public-facing copy (README, docs, manifest strings, release notes, commit messages). Rewrite with colons, semicolons, periods or parentheses instead.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 126 lines · 1,813 tokens per session scan A 87fa984c98b7
planhat-mcp AGENTS.md is an instructions file published in the GitHub repository da-troll/planhat-mcp (0 stars, last pushed today), licensed MIT. It adds 1,813 tokens to every session, about $0.0091 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
ebay-mcp AGENTS.md
Instructions for YosefHayim/ebay-mcp, covering agents.md, what this is, instruction sources, validation commands and module map (src/).
ebay-mcp copilot-instructions.md
Instructions for YosefHayim/ebay-mcp, covering github copilot instructions, core expectations, schema and api patterns, documentation and tests and validation.
ebay-mcp CLAUDE.md
Instructions for YosefHayim/ebay-mcp, covering claude.md and claude-specific notes.
mcp-hubspot CLAUDE.md
Claude Code instructions for baryhuang/mcp-hubspot, covering claude.md, project overview, development commands, installation and setup and install in development mode.
attio-simple-mcp-server CLAUDE.md
Claude Code instructions for mlemos/attio-simple-mcp-server, covering claude.md, project overview, technology stack, available tools (14) and key files.
blockrun-mcp AGENTS.md
AGENTS.md instructions for BlockRunAI/blockrun-mcp, covering blockrun mcp, commands, project structure, key dependencies and install in codex.