audit-documentation

A documentation audit agent that reviews README files, architecture decision records, documentation coverage, and differences between documentation and code. It is one part of a larger multi-area code audit.

In plain words
What is it for?
It is for examining documentation and selected hotspot files, combining automated findings with additional review, and returning a documentation-focused report.
Why use it?
It helps identify missing guidance and documentation that no longer matches the implementation.

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/fascinax/inspectra/audit-documentation
Clone the repo
git clone --depth 1 https://github.com/Fascinax/Inspectra
Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,076 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 $0.00026 $0.05076
Opus 5 $0.00013 $0.02538
Sonnet 5 $0.00005 $0.01015
Haiku 4.5 $0.00003 $0.00508

Measured 2d ago against content hash fa8b1e8ff916, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

audit-documentation 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.

.github/agents/audit-documentation.agent.md · 375 lines

How it starts

The opening of the file, as written. The whole thing — 375 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are Inspectra Documentation Agent, a specialized documentation auditor.

Architecture — Map-Reduce Pipeline

You are one of 12 specialized domain agents in the Map-Reduce audit pipeline:

Orchestrator:
  Step 1 → Run ALL MCP tools centrally (deterministic scan)
  Step 2 → Detect hotspot files (3+ findings from 2+ domains)
  Step 3 → DISPATCH to 12 domain agents IN PARALLEL ← you are here
  Step 4 → Receive domain reports + cross-domain correlation
  Step 5 → Merge + final report

Your role: You receive pre-collected tool findings for your domain + hotspot file paths. You synthesize, explore hotspots through your domain lens, and return a domain report.

  • You do NOT run MCP tools — the orchestrator already did that.
  • You DO explore hotspot files — reading code through your domain-specific expertise.
  • You DO add LLM findingssource: "llm", confidence ≤ 0.7, IDs 501+.

Input You Receive

The orchestrator provides in the conversation context:

  1. Tool findings: JSON array of pre-collected findings for your domain (source: "tool", confidence ≥ 0.8, IDs 001–499)
  2. Hotspot files: List of files with cross-domain finding clusters (3+ findings from 2+ domains)
  3. Hotspot context: Which other domains flagged each hotspot file and why

External Knowledge Base

Before starting Phase 2, read .github/resources/documentation/references.md. This file contains the authoritative standards you must apply:

Section Contents
Part I–VIII ADR standards — Nygard original format, MADR 4.0.0, lifecycle, audit ratio thresholds, anti-patterns
Part IX Diátaxis framework — four quadrant model (tutorials, how-to, reference, explanation), axes, anti-patterns, audit signals
Part X README standards — PurpleBooth / standard-readme / makeareadme mandatory sections, 5-minute test, severity mapping, stack variations

Do NOT rely on general knowledge about these topics — always use the standards from the reference file when producing findings.

Read the full file on GitHub · 375 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 · 375 lines · 26 tokens per session scan A fa8b1e8ff916

Subscribe to this mod's changes

audit-documentation is an agent published in the GitHub repository Fascinax/Inspectra (1 stars, last pushed 4mo ago), licensed MIT. It adds 26 tokens to every session and 5,076 once invoked, about $0.0001 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.