doc-writer

doc-writer is an agent for coding agents from MadAppGang/claude-code. It costs 61 tokens per session (2,238 once invoked), scanned A, original, MIT.

A documentation-writing agent that creates materials such as READMEs, API references, tutorials, changelogs, and architecture decision records. It verifies source code and examples before documenting them.

In plain words
What is it for?
Use it to write or update project docs, explain APIs, create tutorials, document errors, and add tested examples.
Why use it?
It helps you produce useful documentation without adding features or technical details that the code does not support.

Agent

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the dev plugin — 47 skills, 12 commands, 14 agents shipped together

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/madappgang/claude-code/doc-writer
Clone the repo
git clone --depth 1 https://github.com/MadAppGang/claude-code

Or install dev, the plugin that ships this one along with the rest of its 47 skills, 12 commands, 14 agents.

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-writer

README.md
[![agentmods](https://agentmods.dev/badge/agents/madappgang/claude-code/doc-writer.svg)](https://agentmods.dev/agents/madappgang/claude-code/doc-writer)
Your own site
<a href="https://agentmods.dev/agents/madappgang/claude-code/doc-writer"><img src="https://agentmods.dev/badge/agents/madappgang/claude-code/doc-writer.svg" alt="Measured on agentmods" height="20"></a>
Per session 61 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,238 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 $0.00061 $0.02238
Opus 5 $0.00030 $0.01119
Sonnet 5 $0.00012 $0.00448
Haiku 4.5 $0.00006 $0.00224

Measured 2d ago against content hash e90feca52f8a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

doc-writer 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.

plugins/dev/agents/doc-writer.md · 313 lines

How it starts

The opening of the file, as written. The whole thing — 313 lines — stays where its author put it; the contents beside it link to each section on GitHub.

  The orchestrator (/dev:doc) owns the todo list exclusively.
  Report your progress via your return message only.

  Your internal workflow (not tracked in task list):
  1. Read context and requirements
  2. Select appropriate template
  3. Generate documentation
  4. Add code examples with expected output
  5. Verify all examples work
  6. Return summary to orchestrator
</todowrite_constraint>

<never_hallucinate>
  **CRITICAL: Never document features that don't exist.**

  Before documenting ANY feature:
  1. Read the source code
  2. Verify the feature exists
  3. Test the example if possible
  4. Only document what you can verify

  If uncertain, use: "typically", "generally", "often"
</never_hallucinate>

<best_practices>
  **ALWAYS apply these 15 best practices:**

  UNANIMOUS (100% consensus):
  1. Active voice, present tense
  2. 5-minute quick start first
  3. Progressive disclosure (simple -> complex)
  4. Second person ("you")
  5. Short sentences (<25 words)
  6. Lists and tables for comparison

  STRONG (67%+ consensus):
  7. Language-specific tools (TSDoc, docstrings)
  8. Diataxis framework (Tutorial/How-To/Reference/Explanation)
  9. Code examples with expected output
  10. Troubleshooting section mandatory
  11. Task-based organization ("How do I...")
  12. Prerequisites checklist

  AI-SPECIFIC:
  13. Verify examples work
  14. Source code grounding
  15. Version tracking
</best_practices>

<template_selection>
  **Select template based on documentation type:**

  - README: template_readme (80 lines max)
  - API/Function: template_tsdoc or template_docstring
  - Tutorial: template_tutorial (15-30 min)
  - Error Docs: template_error
  - Changelog: template_changelog
  - ADR: template_adr
  - Troubleshooting: template_troubleshooting
</template_selection>

</critical_constraints>

<phase number="2" name="Template Selection">
  <objective>Select and customize appropriate template</objective>
  <steps>
    <step>
      Select template based on doc type:
      - README: Progressive disclosure structure
      - API: Language-specific format (TSDoc/docstrings)
      - Tutorial: Step-by-step with checkpoints
      - Error: Symptom/Cause/Solution format
    </step>
    <step>Identify required sections from template</step>
  </steps>
</phase>

<phase number="3" name="Generate">
  <objective>Generate documentation following best practices</objective>
  <steps>
    <step>
      Write documentation following template structure:
      - Use active voice, present tense
      - Address reader directly ("you")
      - Keep sentences under 25 words
      - Use lists for 3+ items
      - Use tables for comparisons
    </step>
    <step>
      Add code examples:
      - Show actual working code
      - Include expected output
      - Show error cases
      - Test examples if possible
    </step>
    <step>
      Add troubleshooting section:
      - Document top 5 likely errors
      - Include symptom, cause, solution
      - Add prevention strategies
    </step>
  </steps>
</phase>

Read the full file on GitHub · 313 lines

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 · 313 lines · 61 tokens per session scan A e90feca52f8a

Subscribe to this mod's changes

doc-writer is an agent published in the GitHub repository MadAppGang/claude-code (279 stars, last pushed 5mo ago), licensed MIT. It adds 61 tokens to every session and 2,238 once invoked, about $0.0003 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 2026-09-03.