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/alcor6502/codifier-mcp/claude-mdgit clone --depth 1 https://github.com/alcor6502/codifier-mcpWhat 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 | $0.01731 | $0.01731 |
| Opus 5 | $0.00865 | $0.00865 |
| Sonnet 5 | $0.00346 | $0.00346 |
| Haiku 4.5 | $0.00173 | $0.00173 |
Grade A, and why
codifier-mcp 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 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.
How it starts
The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
codifier-mcp — working notes for Claude Code
Rule registry over SQLite, served as an MCP server. Twin of archivist-mcp;
both pin mcp-common-engine to a tag. Two servers in one process — the MCP
surface (server.py) and the admin page (web.py) — over one engine
(rules.py), because two processes on one SQLite database do not share the lock
that makes a multi-statement transaction atomic.
English in this repo: code, docstrings, errors, README, commit messages, tags.
Before you touch anything
VERSIONlives inrules.py, notserver.py. The engine is what the suites import; the server imports the engine. A grep againstserver.pyfinds nothing and reads like "no version".- A number is never written twice. Versions, counts, caps: read them from the
constant or count them. Where a copy is unavoidable, a check must compare the
two —
test_surface.pyis full of them. - One change at a time, in the working tree too. The bill comes at commit: two changes already mixed in the same files cannot be separated afterwards.
git addnamed files. Never-A. Readgit status --shortline by line and ask whether each one is yours; a line you do not recognise is not.
Running the suites
export PYTHONPATH=<mcp-common-engine, EXTRACTED FROM THE PINNED TAG>
python3.12 test_schema.py # DDL: triggers, constraints, schema generation
python3.12 test_registry.py # projects.txt, the router, its refusals
python3.12 test_collaudo.py # the engine, refusals included
python3.12 test_surface.py # the seam: image, template, manuals, README
python3.12 test_crash.py # SIGKILL mid-transaction
No network, no FastMCP, no Docker.
- Take the engine from the pinned tag, not from a sibling clone. That clone is shared with the twin and drifts.
- Read the exit code. A suite that dies halfway still prints hundreds of
greens; a
| tailthrows away the only signal that says so. - A bench built with
mktemp -dneeds the non-Python files too, and.github/workflows/. Without that directorytest_surface.pydoes not go red — it dies withFileNotFoundErroronbuild.yml, after printing hundreds of passes. Compare the sha256 ofserver.pyandrules.pybetween source and bench: a stale bench answers for code you are not running.
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.
- 2d ago First seen · 129 lines · 1,731 tokens per session scan A 2dc1f96d1898
codifier-mcp CLAUDE.md is an instructions file published in the GitHub repository alcor6502/codifier-mcp (0 stars, last pushed 3d ago), licensed MIT. It adds 1,731 tokens to every session, about $0.0087 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.