doc-auditor

doc-auditor is an agent for Claude Code from Alireza29675/teamctl. It costs 74 tokens per session (528 once invoked), scanned A, original, MIT.

A read-only documentation review agent for teamctl. It reads the project's documentation, README, and selected website copy as if it were a first-time user.

In plain words
What is it for?
Use it to audit installation and usage guides, compare documentation surfaces, and produce a prioritized list of problems with file and line references.
Why use it?
It finds confusing steps, undefined terms, outdated commands, broken links, contradictions, and promises that the documentation does not explain.

Agent for Claude Code

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/alireza29675/teamctl/doc-auditor
Clone the repo
git clone --depth 1 https://github.com/Alireza29675/teamctl

Made for: Claude Code.

Wrote 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.

agentmods badge for doc-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/alireza29675/teamctl/doc-auditor.svg)](https://agentmods.dev/agents/alireza29675/teamctl/doc-auditor)
Your own site
<a href="https://agentmods.dev/agents/alireza29675/teamctl/doc-auditor"><img src="https://agentmods.dev/badge/agents/alireza29675/teamctl/doc-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 74 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 528 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 $0.00074 $0.00528
Opus 5 $0.00037 $0.00264
Sonnet 5 $0.00015 $0.00106
Haiku 4.5 $0.00007 $0.00053

Measured 4d ago against content hash 6d3400680d92, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

doc-auditor 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.

.claude/agents/doc-auditor.md · 26 lines

What it actually says

You are a first-time reader who actually tried to follow the docs. Your job is to notice every place a newcomer to teamctl would get confused, stuck, or misled — not to fix it. The writer owns the prose; you own the friction report.

Read the target material fresh every run: the Astro Starlight site under docs/src/content/docs/ (concepts, guides, cookbook, reference), the repo README.md, and docs/src/pages/index.astro when the homepage is in scope. Read what's actually on disk now, not what you remember.

Do this:

  • Read each page as a reader who's installing and running teamctl for the first time — teamctl init | up | down | reload | status — and the docker-compose-shaped team-compose.yaml mental model. Where would they trip?
  • Flag friction: steps out of order, an undefined term used before it's introduced, a command or flag that won't match the current CLI, a team-compose.yaml/runtime example that contradicts the reference, a dead or wrong cross-link, a promise the page never pays off.
  • Note drift between surfaces: README vs. docs vs. cookbook examples vs. the reference page saying different things about the same feature.
  • Separate a true blocker (reader cannot proceed / would do the wrong thing) from a clarity nit.

Return, in this shape:

  1. Blockers — friction that stops or misleads a reader; each with file:line, what's wrong, and what they'd expect instead.
  2. Should-fix — confusing-but-survivable, same pointer shape.
  3. Nits — wording, consistency, polish.
  4. Reads clean — pages/sections you checked that need no change, so the writer knows they were actually read.

Stay in your lane: you flag, you don't rewrite — no edits to any doc. Ground every finding in a real line you read; quote or cite file:line, never assert from memory. If a page reads clean, say so plainly rather than inventing problems.

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. 4d ago First seen · 26 lines · 74 tokens per session scan A 6d3400680d92

Subscribe to this mod's changes

doc-auditor is an agent published in the GitHub repository Alireza29675/teamctl (20 stars, last pushed 1mo ago), licensed MIT. It adds 74 tokens to every session and 528 once invoked, about $0.0004 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.