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/appautomaton/automaton/agents-mdgit clone --depth 1 https://github.com/appautomaton/automatonWhat 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 | $0.00844 | $0.00844 |
| Opus 5 | $0.00422 | $0.00422 |
| Sonnet 5 | $0.00169 | $0.00169 |
| Haiku 4.5 | $0.00084 | $0.00084 |
Grade A, and why
automaton 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md: automaton
Portable, stage-gated agentic-AI harness for Claude Code, Codex, and OpenCode.
Architecture
- Copy-based install. Skills and shared runtime files are inspectable after install.
_shared/installs once under.agent/.automaton/where host skills can reference it (seedocs/design-decisions.mdDD-001). Install writes a receipt (.agent/.automaton/state/install-manifest.json) recording files, created directories, and merged config fragments per host, so uninstall and upgrade act on exactly what this project received (DD-011). - Skills are pure markdown.
SKILL.md+references/, plusrole-sources/in auto-execute for the dispatch roles. Shared helper scripts live inskills/_shared/scripts/. - Three hosts. Claude, Codex, and OpenCode. Each wires startup context through host hooks/plugins and maps subagent tools.
- Five lifecycle stages plus resume. Stages:
frame,plan,execute,verify,verified.resumere-enters from durable state. Prerequisites are enforced inruntime/lib/contracts-data.json. - No mandatory nested invocation. Skills hand off through durable artifacts. Clean same-session continuation is allowed when the lifecycle contract says it is safe.
Layout
bin/ CLI (install, status, validate, context)
hosts/ Host adapters (Claude, Codex, OpenCode)
lib/ CLI + install library (re-exports runtime)
runtime/ Installed into target projects (state, context, validation)
skills/ Skill sources, edit here, not in installed copies
_shared/ Shared refs + scripts (installed) and authoring guides (not installed)
auto-*/ Individual skills
tests/ node --test tests/*.test.mjs
docs/ Design rationale (not shipped in npm package)
Commands
npm test # all tests
node --test tests/contracts.test.mjs # targeted test
node bin/automaton.mjs install --claude .
node bin/automaton.mjs status
node bin/automaton.mjs validate
node bin/automaton.mjs context frame
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 · 56 lines · 844 tokens per session scan A 66c863a3246e
automaton AGENTS.md is an instructions file published in the GitHub repository appautomaton/automaton (21 stars, last pushed 16d ago), licensed MIT. It adds 844 tokens to every session, about $0.0042 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 2026-08-30.
Other instructions, from other repositories
plannotator AGENTS.md
AGENTS.md instructions for backnotprop/plannotator, covering plannotator, project structure, server runtimes, installation and environment variables.
golembot CLAUDE.md
Instructions for 0xranx/golembot, covering golembot — project conventions for claude code, architecture hard constraints, things you must never do, interface change rules and file responsibility boundaries.
solo AGENTS.md
Instructions for solo-agent/solo, covering project testing rules and project service lifecycle rules.
OpenCode-GraphAgent CLAUDE.md
Instructions for LeXwDeX/OpenCode-GraphAgent, covering claude.md, what this is, commands, quality gates and build & codegen.
superset CLAUDE.md
Claude Code instructions for superset-sh/superset, a project described as: Superset is an agentic IDE to orchestrate 100+ coding agents in parallel. Run any agent with your own subscription.
awesome-copilot-id AGENTS.md
Instructions for GulajavaMinistudio/awesome-copilot-id, covering communication, explanation and documentation, markdown formatting, user communication style and workflow & methodology.