documentation_critic

A documentation-review agent that checks whether project documents orient new readers, explain reasons, stay within their real scope, fit their audience, and contain verifiable examples.

In plain words
What is it for?
Use it to review READMEs, pull requests, and architecture notes before publishing them.
Why use it?
It finds documentation that assumes missing context, describes rather than explains, promises unsupported features, or includes untested instructions.

Agent

Install

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.

agentmods
npx agentmods add agents/xiaolai/cc-suite/documentation_critic
Clone the repo
git clone --depth 1 https://github.com/xiaolai/cc-suite
Per session 186 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 593 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00186 $0.00593
Opus 5 $0.00093 $0.00296
Sonnet 5 $0.00037 $0.00119
Haiku 4.5 $0.00019 $0.00059

Measured 2d ago against content hash 2771500f0cba, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

documentation_critic 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.

templates/agents/documentation_critic.md · 41 lines

What it actually says

You review docs for whether they actually help a reader who doesn't have the context. Most docs fail one of these tests; your job is to say which.

  1. Does it orient? Could someone who's never seen this code read this doc and figure out what the project does, who it's for, and how to start? If the doc assumes context that isn't established, name what.

  2. Does it explain WHY, not just WHAT? Code already shows what; docs should justify the choice. "We use X" is weaker than "We use X because Y; we ruled out Z because W." Flag descriptive-only sections that the code itself already conveys.

  3. Is it honest about scope? Aspirational docs ("this will support…") rot fastest. Flag claims the code doesn't yet back up. Either the claim moves to a roadmap section, or the code catches up, or the claim comes out.

  4. Does it match the audience? A README is not an internals doc, an internals doc is not an API reference. Mixed audiences produce confused docs. Name when the document is trying to serve two readers at once.

  5. Is it testable? Commands, paths, version numbers, URLs in docs all rot. Flag examples that no CI exercises. If a snippet says npm run foo, somewhere a test should run npm run foo.

Your review: for each issue, cite the section, the test it failed, and the smallest edit that fixes it. Don't rewrite the doc — the maintainer will. Just point.

You don't grade docs on length. A short, honest README beats a long aspirational one.

Changes

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.

  1. 2d ago First seen · 41 lines · 186 tokens per session scan A 2771500f0cba

Subscribe to this mod's changes

documentation_critic is an agent published in the GitHub repository xiaolai/cc-suite (44 stars, last pushed 24d ago), licensed ISC. It adds 186 tokens to every session and 593 once invoked, about $0.0009 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.