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/jakubsuplicki/codument/claude-mdgit clone --depth 1 https://github.com/jakubsuplicki/codumentWrote 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/jakubsuplicki/codument/claude-md)<a href="https://agentmods.dev/instructions/jakubsuplicki/codument/claude-md"><img src="https://agentmods.dev/badge/instructions/jakubsuplicki/codument/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 | $0.04678 | $0.04678 |
| Opus 5 | $0.02339 | $0.02339 |
| Sonnet 5 | $0.00936 | $0.00936 |
| Haiku 4.5 | $0.00468 | $0.00468 |
Grade A, and why
codument 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 5d 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.
This is a copy
92% identical to codument AGENTS.md — 44 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codument Claude Compatibility
Shared agent guidance lives in AGENTS.md. Treat that file as the canonical project contract; this file remains for Claude-specific tooling compatibility.
Claude Compatibility
Shared agent guidance lives in AGENTS.md. Follow that file as the canonical Codument workflow contract.
Codument Delivery Workflow
Core loop
Use Codument as the durable control plane for agent-led engineering work:
- Grill the request against existing docs, code, ADRs, and project language.
- Plan the feature in docs before changing source code.
- Wait for explicit user approval before implementation.
- Implement one planned step at a time.
- Build the strongest practical feedback loop, preferring red-green-refactor when it fits.
- Update the mapped docs +
docs/.registry.jsonas part of the same step: materialize each new source file withcodument map materialize; when a contract changed, update its doc at intent altitude (a move that changed no contract owes nothing at all — the gate reports it and never blocks, so there is no signature to write). Never a junk mirror edit to clear the gate. - Review the diff against the approved plan, tests, docs, and architecture.
- Commit focused work with a conventional commit, authored as the user with no AI
Co-Authored-Bytrailer. - Move to the next unchecked step.
If a codument command's quoted argument comes back refused as several arguments — --reason "one two three" rejected as three — the launcher split it before codument saw argv, so no quoting fixes it. Run the CLI a different way for the rest of the session: npx codument …, or node node_modules/codument/dist/cli.js …. Seen with bunx on Windows.
Quality bar
Aim for the best-effort, durable solution, not the first plausible one. Before calling a plan or a step done, zoom out and check it adversarially — where is this half-baked, what did I assume, what would break it. Resolve issues yourself; pull the user in only for a genuinely load-bearing, unconfirmed call (the assumption gate below), not for work that should just happen.
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.
- 5d ago First seen · 143 lines · 4,678 tokens per session scan A b137e4146022
codument CLAUDE.md is an instructions file published in the GitHub repository jakubsuplicki/codument (47 stars, last pushed 14d ago), licensed Apache-2.0. It adds 4,678 tokens to every session, about $0.0234 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to codument AGENTS.md, differing in 44 lines, and is treated as a copy.
Other instructions, from other repositories
vibe-coding-prompt-template backend.instructions.md
Instructions for KhazP/vibe-coding-prompt-template: Read AGENTS.md, agentdocs/techstack.md, and agentdocs/codepatterns.md.
core copilot-instructions.md
Copilot instructions for asdecided/core: These decisions are already accepted. Do not re-open or contradict them; ask the AsDecided MCP tools (getartifact, searchartifacts) for the full text before proposing a change that touches one.
slideops CLAUDE.md
Claude Code instructions for glukicov/slideops, covering claude.md, what this repository is, commands, the demo deck's citations must still match this repo and the citation mechanism (the core idea).
valuein CLAUDE.md
Claude Code instructions for valuein/valuein, covering claude.md, what this repo is (and isn't), sibling repos — where the actual code lives, commands and with a paid/sp500 token.
living-manual CLAUDE.md
Instructions for dougstanford/living-manual, covering living-manual, start of every session, how work lands — non-negotiable, user's manual (living-manual) and releasing.
sourcebook CLAUDE.md
Instructions for BackendGameSetMatch/sourcebook, covering ai documentation, when to consult docs, how to access, keep docs alive and hard rules from conventions.