interdoc AGENTS.md

A tool that creates and maintains AGENTS.md files—guides that tell coding agents how to work in different parts of a project. It analyses the project structure and can also check or repair the documentation.

In plain words
What is it for?
Use it to generate or update project instructions, review documentation coverage, lint AGENTS.md files, fix stale references, or preview changes without writing files.
Why use it?
It keeps agent instructions aligned with the repository as files and conventions change. It removes the manual work of writing, updating, and checking these guides.

Instructions file for CodexOpenCode

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 instructions/mistakeknot/interdoc/agents-md
Clone the repo
git clone --depth 1 https://github.com/mistakeknot/interdoc

Made for: Codex, OpenCode.

Per session 821 This file is loaded in full into every session.
When invoked 821 The same file — it is already loaded in full.
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 $0.00821 $0.00821
Opus 5 $0.00411 $0.00411
Sonnet 5 $0.00164 $0.00164
Haiku 4.5 $0.00082 $0.00082

Measured 2d ago against content hash cdd2aa3e3abf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

interdoc AGENTS.md 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.

AGENTS.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — interdoc

Recursive AGENTS.md generator with integrated Oracle critique. Analyzes project structure via parallel subagents, consolidates into coherent documentation, and optionally sends to GPT for independent review.

Plugin Type: Claude Code skill plugin Plugin Namespace: interdoc (from interagency-marketplace) Current Version: 5.0.0

Canonical References

  1. PHILOSOPHY.md — direction for ideation and planning decisions.
  2. CLAUDE.md — implementation details, architecture, testing, and release workflow.

Quick Reference

Trigger Mode Scope
"generate AGENTS.md" Generation Full project
"update AGENTS.md" Update Changed directories
"change-set update" Update Git diff only
"doc coverage" Report Coverage stats
"lint AGENTS.md" / "doc lint" Lint Style warnings
"fix stale references" / "interdoc fix" Fix Structural only (no LLM)
"dry run" Any + Preview No writes

Topic Guides

Topic File Covers
Skill & Features agents/skill-and-features.md Skill triggers, key features, interwatch integration, hooks
Development & Versioning agents/development.md Editing behavior, testing, version bumps, marketplace publish, commit workflow
GPT Review agents/gpt-review.md Oracle integration, review classification, helper scripts, troubleshooting
Changelog agents/changelog.md Version history (v4.1.0 through v4.4.3)

Repository Structure

/
├── .claude/
│   └── agents/
│       └── interdocumentarian.md  # Claude subagent for AGENTS.md authoring
├── .claude-plugin/
│   └── plugin.json          # Plugin metadata and version (source of truth)
├── .codex/
│   └── INSTALL.md            # Codex CLI install instructions
├── hooks/
│   ├── check-updates.sh     # Hook script (disabled by default)
│   └── check-commit.sh      # Hook script (disabled by default)
│   ├── git/
│   │   ├── post-commit       # Advisory commit hook
│   │   └── install-post-commit.sh
│   └── tools/
│       └── interdoc-audit.sh  # Coverage + lint helper
├── scripts/
│   ├── drift-fix.sh       # Structural auto-fix (renames, deletions, link fixes)
│   ├── bump-version.sh    # Version management
│   ├── check-versions.sh  # Version consistency checker
│   └── interdoc-generator.sh
├── skills/
│   └── interdoc/
│       └── SKILL.md         # Main skill definition
├── docs/
│   ├── plans/               # Design documents
│   └── TEST_PLAN.md         # Test cases from splinterpeer analysis
├── tests/
│   ├── fixtures/
│   │   └── setup-test-repo.sh  # Test repo scaffolding
│   ├── test-drift-fix-*.sh     # drift-fix.sh test suite
│   └── run-all.sh              # Test runner
├── README.md                # User-facing documentation
├── CLAUDE.md                # Claude-specific settings only
└── AGENTS.md                # This file - cross-AI documentation

Read the full file on GitHub · 74 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 · 74 lines · 821 tokens per session scan A cdd2aa3e3abf

Subscribe to this mod's changes

interdoc AGENTS.md is an instructions file published in the GitHub repository mistakeknot/interdoc (2 stars, last pushed 1mo ago), licensed MIT. It adds 821 tokens to every session, about $0.0041 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.