Borrowing it
Nothing to install: this file belongs to racecraft-lab/racecraft-plugins-public. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/racecraft-lab/racecraft-plugins-public/main/AGENTS.mdgit clone --depth 1 https://github.com/racecraft-lab/racecraft-plugins-publicWrote 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/racecraft-lab/racecraft-plugins-public/agents-md)<a href="https://agentmods.dev/instructions/racecraft-lab/racecraft-plugins-public/agents-md"><img src="https://agentmods.dev/badge/instructions/racecraft-lab/racecraft-plugins-public/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/racecraft-lab/racecraft-plugins-public/agents-md"><img src="https://agentmods.dev/badge/instructions/racecraft-lab/racecraft-plugins-public/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.05791 | $0.05791 |
| Opus 5.5 | $0.02316 | $0.02316 |
| Sonnet 5.5 | $0.01158 | $0.01158 |
| Haiku 4.5 | $0.00579 | $0.00579 |
Grade A, and why
racecraft-plugins-public 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 14d 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 — 332 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Agent Instructions
The shared agent contract for Codex, Claude Code, Gemini, and Copilot. It loads into every session, so it holds durable behavior only: no release history, workflow runbooks, generated-plan exhaust, or long plugin internals. Keep it under 200 lines; put detail in the docs listed under Read When Relevant.
1. Start Here
A public Claude Code and Codex marketplace with two plugins: speckit-pro,
which installs from dist/, and typesafe-jev, which installs from
typesafe-jev/plugin/ with no generated payload. Specs and planning artifacts
are context on demand. Before editing speckit-pro/, tests/speckit-pro/, or
docs-site/, read its scoped AGENTS.md; Codex loads one only when started
inside that directory.
| Area | Where to look |
|---|---|
| Claude skills, agents, hooks | speckit-pro/skills/, speckit-pro/agents/, speckit-pro/hooks/ |
| Codex skills, agents, hooks | speckit-pro/skills/ overlaid by speckit-pro/codex-skills/, speckit-pro/codex-agents/ (TOML), speckit-pro/codex-hooks.json |
| Python runner | speckit-pro/speckit_pro_runner/: gates in gates/, helper ids in helpers/registry.py |
| Tests | tests/speckit-pro/; layers and default selection in suite-manifest.json |
| typesafe-jev | Go source and Go tests in typesafe-jev/cmd/evaluate/; shipped plugin in typesafe-jev/plugin/ |
2. Commands
Run from the repository root (Python 3.11+, Node >= 22.12 and pnpm for docs). A
fresh worktree holds only tracked files; before any docs command, run
pnpm --dir docs-site install --frozen-lockfile.
| Check | Command | CI job (required?) |
|---|---|---|
Quick suite: toolchain, layers 1, 4, 5 (--layer 1, 4, 5, or 6 for one) |
python3 tests/speckit-pro/run-all.py |
none |
CI suite: adds layers 6 and 7 (run-all.py cannot select 7; python3 tests/speckit-pro/run-layer-scripts.py --layer 7 runs it alone) |
python3 scripts/run-ci-suite.py (the one entry point for local and CI runs; worktree-isolated agents must use it). Equivalent raw command, pinned to the script by test-plain-ci-entry-points.py: SPECKIT_SKIP_TOOLCHAIN_CHECK=1 GIT_CONFIG_GLOBAL=/dev/null GIT_CONFIG_SYSTEM=/dev/null GIT_CONFIG_NOSYSTEM=1 PYTHONPATH=speckit-pro python3 -m speckit_pro_runner < tests/speckit-pro/unit/fixtures/runner-gates/requests/run-ci-suite.json |
test (yes, via validate-plugins) |
| typesafe-jev Go checks | python3 scripts/check-go-module.py check |
go (yes, via validate-plugins) |
| Generated-artifact drift; commit first, since any uncommitted change under its paths fails it | python3 scripts/refresh-release-artifacts.py --check |
artifact-consistency (yes, via validate-plugins) |
| PR title | python3 scripts/check-pr-title.py '<title>' (the one entry point for local and CI runs). Equivalent raw command, pinned to the script by test-plain-ci-entry-points.py: TITLE='<title>' PYTHONPATH=speckit-pro python3 -m speckit_pro_runner < tests/speckit-pro/unit/fixtures/runner-gates/requests/validate-pr-title-live.json |
validate-pr-title (yes) |
| Release-note fence | PR_TITLE='<title>' PR_BODY='<body>' PR_LABELS_JSON='[]' python3 scripts/compose-release-notes.py --validate-pr |
validate-release-note (yes) |
| Docs, reference mode: reference inputs changed | pnpm --dir docs-site reference:check, then pnpm --dir docs-site validate:quality |
validate-docs (no) |
Docs, full mode: docs-site/, the artifact gallery, or a docs contract file changed (scripts/classify-docs-validation.py) |
pnpm --dir docs-site exec playwright install --with-deps chromium once, then pnpm --dir docs-site validate |
validate-docs (no) |
| Container preflight: Linux containers rerun the suite when runner, test, or workflow paths change | CI only; its extra requests (LINUX_REQUESTS in tests/speckit-pro/run-container-preflight.py) also run locally |
container-preflight-linux-amd64, -arm64 (yes) |
Python lint: ruff F, B, and BLE rules (ruff.toml); mypy over the mypy.ini allowlist (add a module once it passes) |
In a virtual environment, python3 scripts/run-python-lint.py install ruff, then run ruff; the same for mypy |
python-lint (no; built to be required), mypy-ratchet (no) |
| Workflow lint | actionlint at the version pinned in pr-checks.yml, from the repository root. The CI installer (scripts/install-actionlint.py) fetches a Linux amd64 binary only |
validate-workflows (no; it also checks release-PR ancestry, CI only) |
Ripwire advisory: --arch, --quality-delta from the merge-base, and --doc-drift, reported in the job summary |
CI only; scripts/install-ripwire.py pins one Linux release by SHA-256. Locally, run the ripwire . commands in section 5 |
ripwire-advisory (no; advisory, never required) |
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.
- 14d ago Changed · -1 lines · -8 tokens per session 5e74f854a3ea
- 16d ago Changed · +42 tokens per session fb3bfaaac71f
- 22d ago Changed · +48 tokens per session d06158d1ff7a
- 23d ago Changed · +8 lines · +149 tokens per session e75c097db249
- 25d ago Changed · +38 lines · +759 tokens per session 3f5d1c3a9a13
- 26d ago Changed · +55 lines · +1,198 tokens per session 8732aea9f946
- 27d ago Changed · +8 lines · +151 tokens per session 04cfd01f111a
- 1mo ago Changed · +18 lines · +288 tokens per session cee934ca73c0
- 1mo ago Changed · +6 lines · +123 tokens per session 63ec9fc1385d
- 1mo ago First seen · 200 lines · 3,041 tokens per session scan A 4a8d0947ca1a
racecraft-plugins-public AGENTS.md is an instructions file published in the GitHub repository racecraft-lab/racecraft-plugins-public (5 stars, last pushed 6d ago), licensed MIT. It adds 5,791 tokens to every session, about $0.0232 per session on Opus 5.5. 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.