agentsmd AGENTS.md

Repository instructions for developing an agents.md project, including its Node.js setup, build commands, tests, and coding conventions. An agents.md file gives coding agents rules for working in a project.

In plain words
What is it for?
Installing dependencies, building the project and GitHub Action, running the full verification gate, and following its TypeScript and commit rules.
Why use it?
They define the checks and conventions needed for reliable, repeatable changes, including deterministic output that does not vary between runs.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/daichunghy/agentsmd/agents-md
Clone the repo
git clone --depth 1 https://github.com/daichunghy/agentsmd

Made for: Codex, OpenCode.

Per session 258 This file is loaded in full into every session.
When invoked 258 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00258 $0.00258
Opus 5 $0.00129 $0.00129
Sonnet 5 $0.00052 $0.00052
Haiku 4.5 $0.00026 $0.00026

Measured 2d ago against content hash b8aa82a3e482, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentsmd 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 2d 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.

AGENTS.md · 29 lines

What it actually says

AGENTS.md

Setup

Run npm install once; Node >= 18.

Build

npm run build compiles src/ to dist/. npm run build:action bundles the GitHub Action entry to action-dist/index.js (committed).

Test

npm run verify is the full gate: typecheck, build, action bundle, tests (including golden fixtures, determinism, text/JSON parity, and process tests). Run it before considering any change complete.

Conventions

  • TypeScript strict mode; no any in src/.
  • The CLI keeps zero runtime dependencies; Action-only dependencies are inlined into action-dist at build time.
  • Core logic is pure over the injected FileReader; tests use MemFs fixture trees from fixtures/.
  • Lint/score output must stay deterministic: no timestamps, no absolute paths, canonical JSON with sorted keys.
  • Conventional commits (feat:, fix:, test:, chore:, docs:).
  • Design authority: docs/superpowers/specs/2026-08-22-agentsmd-mvp-design.md.
Changes

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.

  1. 2d ago First seen · 29 lines · 258 tokens per session scan A b8aa82a3e482

Subscribe to this mod's changes

agentsmd AGENTS.md is an instructions file published in the GitHub repository daichunghy/agentsmd (1 stars, last pushed 4d ago), licensed MIT. It adds 258 tokens to every session, about $0.0013 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.