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/blackaxgit/agent-mcp-hub/agents-mdgit clone --depth 1 https://github.com/blackaxgit/agent-mcp-hubWrote 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/blackaxgit/agent-mcp-hub/agents-md)<a href="https://agentmods.dev/instructions/blackaxgit/agent-mcp-hub/agents-md"><img src="https://agentmods.dev/badge/instructions/blackaxgit/agent-mcp-hub/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 | $0.06484 | $0.06484 |
| Opus 5 | $0.03242 | $0.03242 |
| Sonnet 5 | $0.01297 | $0.01297 |
| Haiku 4.5 | $0.00648 | $0.00648 |
Grade D, and why
agent-mcp-hub AGENTS.md scanned grade D with 2 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 4d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **`agy` does not stream and leaves state behind.** Its entire answer is flushed at once on completion (measured), so the idle timer is its only hang protection — see the timeouts section of `README.md`. Each `--new-pro Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **`prebuild: rm -rf dist` is load-bearing — do not remove it.** `tsc` does not clean `dist/`, so files whose sources were deleted linger and get **published**. v0.5.0 shipped dead `dist/http.js` + `dist/httpServer.js` How it starts
The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
AI coding assistant context for this repository.
AGENTS.mdis a tool-agnostic context file read natively by many AI coding tools (Codex, Cursor, Copilot, Gemini CLI, Windsurf, …). Claude Code does not read it natively — Claude users should add@AGENTS.mdto aCLAUDE.md.
Command sources: package.json scripts are authoritative — CI runs the same ones. No Makefile/Taskfile.
Local overlay (read the nearer file when work touches those paths):
| Path | Read when |
|---|---|
.github/workflows/AGENTS.md |
Touching CI, the release pipeline, or anything that publishes. Those rules are load-bearing and break publishing silently if got wrong. |
Overview
- Purpose: One MCP server that bridges multiple CLI coding agents — Codex, Cursor, OpenCode, Claude, and Antigravity (
agy) — into any MCP client. - Owner: TODO: team / maintainer
- Type: app — a published npm package (
agent-mcp-hub, public on npm) - Stack: Node ESM (
"type": "module"), TypeScript,engines.node >= 22, npm (package-lock.json) - Runtime deps:
@modelcontextprotocol/sdk,zod. Everything else is dev-only. - Transport: stdio only. The HTTP transport and Docker packaging were removed in v0.5.0 — a container cannot see the caller's repo path or reuse their CLI logins, which broke the product contract.
- Distribution: npm + the MCP Registry (
server.json). Released automatically — seePUBLISHING.md.
Commands
npm ci # install from package-lock
npm run dev # run from source via tsx (no build step)
npm run build # prebuild wipes dist/, then tsc -> dist/
npm test # vitest run
npm run test:coverage # vitest + coverage (97% thresholds enforced)
npm run typecheck # tsc -p tsconfig.test.json (covers src + tests)
npm run lint # eslint .
npm run format # prettier --write . (format:check to verify only)
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.
- 4d ago First seen · 229 lines · 6,484 tokens per session scan D 34fccd5a3772
agent-mcp-hub AGENTS.md is an instructions file published in the GitHub repository blackaxgit/agent-mcp-hub (0 stars, last pushed yesterday), licensed MPL-2.0. It adds 6,484 tokens to every session, about $0.0324 per session on Opus 5. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
meta-llm-charter CLAUDE.md
Instructions for entropyvortex/meta-llm-charter, covering meta v3.1 core charter, bias, meta-0, r1 decompose and r2 ask gate.
awesome-reviewers CLAUDE.md
Instructions for baz-scm/awesome-reviewers, covering awesome reviewers — repository guidelines, what this project is, source of truth, site layer and machine interface.
code-context AGENTS.md
AGENTS.md instructions for infino-ai/code-context, covering code-context: notes for ai agents, project overview, repo map, build, test, gates and conventions.
gtm-cheat-codes AGENTS.md
Instructions for zapier/gtm-cheat-codes, covering agents.md and project overview.
Stata-CLI AGENTS.md
Instructions for Utolaris/Stata-CLI, covering agents.md, project structure and working rules.
specgate CLAUDE.md
Claude Code instructions for thanhtung2693/specgate: This file configures Claude Code when it contributes to the SpecGate repository. It is not product guidance for people using SpecGate.