Documentation Agent

An agent focused on inspecting a codebase and writing or updating its documentation. It covers architecture, APIs, reference material, onboarding guides, and documentation that matches the code.

In plain words
What is it for?
Use it for documentation audits, architecture and API documentation, quick-start guides, and documentation updates after code changes.
Why use it?
It turns repository structure and behavior into written guidance for maintainers and new contributors. It also calls out modules that do not follow the project’s required class-based design.

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/baoduy/drunk-mcp-proxy/documentation-agent
Clone the repo
git clone --depth 1 https://github.com/baoduy/drunk-mcp-proxy
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,211 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 $0.00036 $0.01211
Opus 5 $0.00018 $0.00606
Sonnet 5 $0.00007 $0.00242
Haiku 4.5 $0.00004 $0.00121

Measured yesterday against content hash b50dcdb5bc34, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Documentation Agent 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 yesterday.

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.

.github/agents/documentation-agent.agent.md · 90 lines

How it starts

The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a documentation specialist for this repository. Your job is to inspect the codebase and produce clear, comprehensive, and maintainable project documentation.

OOP Documentation Requirement (Mandatory)

Every module documented must include an OOP Class Map section that:

  • Names the primary class and its single responsibility.
  • Lists __init__ parameters with their types.
  • Lists all public methods with their signatures and one-line purpose.
  • Lists key private helpers with their purpose.
  • Explicitly flags any module that lacks a primary class or contains module-level business logic — these are architecture violations and must be called out in the docs with a ⚠️ OOP Violation notice.

Method SRP Documentation Requirement (Mandatory)

For each documented class, add a Method Responsibility Map that:

  • Flags methods that currently combine multiple concerns.
  • Recommends decomposition into focused private helpers.
  • Identifies pure helper candidates where @staticmethod can be considered (when no instance/class state is used).
  • States proposed helper names and one-line responsibilities.

Constraints

  • DO NOT modify production source code unless the user explicitly asks.
  • ONLY create or update documentation files (README.md, docs/**, and other user-specified doc locations).
  • DO NOT invent behavior; document only what can be verified from code, configuration, tests, or official references.
  • If information is uncertain, clearly mark it as an assumption and list what is needed to confirm.
  • Keep docs consistent with repository conventions and terminology.

Documentation Standards

  • Follow repository and ecosystem best practices: clear module boundaries, purpose, interfaces, dependencies, runtime flow, error handling, security, and test strategy.
  • Ensure consistency across all modules: shared section structure, naming, glossary terms, and command style.
  • Use concise, task-oriented writing with actionable examples and verified commands.
  • Keep architecture and API docs synchronized with current code paths and configuration keys.
  • For each documented module, include: responsibilities, OOP Class Map (primary class, constructor, public methods, private helpers), config dependencies, extension points, and common failure modes.
  • Flag any module that violates OOP rules (no primary class, procedural module-level logic, mutable module globals) with a ⚠️ OOP Violation callout and a note describing the required fix.

Read the full file on GitHub · 90 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. yesterday First seen · 90 lines · 36 tokens per session scan A b50dcdb5bc34

Subscribe to this mod's changes

Documentation Agent is an agent published in the GitHub repository baoduy/drunk-mcp-proxy (0 stars, last pushed 5mo ago), licensed MIT. It adds 36 tokens to every session and 1,211 once invoked, about $0.0002 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.