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/lab2a/metalworks/claude-mdgit clone --depth 1 https://github.com/Lab2A/metalworksWrote 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/lab2a/metalworks/claude-md)<a href="https://agentmods.dev/instructions/lab2a/metalworks/claude-md"><img src="https://agentmods.dev/badge/instructions/lab2a/metalworks/claude-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 | $0.00554 | $0.00554 |
| Opus 5 | $0.00277 | $0.00277 |
| Sonnet 5 | $0.00111 | $0.00111 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
metalworks CLAUDE.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 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.
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 — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — metalworks
Conventions for working on metalworks with Claude Code. CONTRIBUTING.md is the full guide; docs/architecture.md is the why.
Before opening a PR
Run /pr-ready — it runs the gate, the contract-drift check CI doesn't, and walks the
parity / docs / changelog checks. The gate (what CI enforces):
uv run ruff check . && uv run ruff format --check . && uv run pyright && uv run pytest -q
Non-negotiables
- Contract-first, four-surface parity. A primitive lives on the Python facade (
client.py), the CLI (cli/__init__.py), the MCP server (mcp/tools.py+server.py+ the_TOOL_WRAPPERStuple), and a Claude Code skill (plugin/skills/). Change one → change all four, plus the docs. - Registry lockstep. A new
contractmodel goes inscripts/gen_ts_types.pyMODELSandcontract/__init__.py__all__; regenerate (python scripts/gen_ts_types.py) and committs/contract.ts+ thecontract/schema/*.json. CI doesn't run--check—/pr-readydoes. Additive only below 1.0 (new fields defaulted; old payloads still validate). - No-cite-no-claim. Every claim resolves to a real quote (
EvidenceRef→report.evidence); un-grounded output is dropped, never invented. - Lean core, lazy extras. Provider SDKs are lazy-imported inside the function that uses them,
behind an extra;
import metalworksstays free. No env reads or client construction at import time. - Offline tests.
pytest-socketblocks the network; useFakeChatModel/FakeEmbedding/MemoryStores. Mark real-network testsnetwork. - Decisions are deterministic. Verdicts / severity / demand bands are pure functions; the LLM writes only the human-facing rationale.
House style
Match the surrounding code. Ruff (line length 100) and pyright strict over src/ are the law; keep
new code green on both. Skills live in plugin/skills/ (product) and .claude/skills/ (repo dev
tools like /pr-ready).
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 · 38 lines · 554 tokens per session scan A 4fceb3d7a38d
metalworks CLAUDE.md is an instructions file published in the GitHub repository Lab2A/metalworks (6 stars, last pushed 2mo ago), licensed MIT. It adds 554 tokens to every session, about $0.0028 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
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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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 oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.