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.
npx agentmods add instructions/nicograef/handbook/agents-mdgit clone --depth 1 https://github.com/nicograef/handbookWrote 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.
[](https://agentmods.dev/instructions/nicograef/handbook/agents-md)<a href="https://agentmods.dev/instructions/nicograef/handbook/agents-md"><img src="https://agentmods.dev/badge/instructions/nicograef/handbook/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01532 | $0.01532 |
| Opus 5 | $0.00766 | $0.00766 |
| Sonnet 5 | $0.00306 | $0.00306 |
| Haiku 4.5 | $0.00153 | $0.00153 |
Grade A, and why
handbook 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 today.
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.
How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agents
| Surface | Role |
|---|---|
AGENTS.md |
The single canonical instruction set; read by the Copilot coding agent, code review on github.com, and Copilot CLI (not Chat) |
CLAUDE.md |
Loads this file for Claude Code via @AGENTS.md |
.claude/rules/*.md |
The only path-scoped conventions surface; Claude Code loads it on matching files |
agents symlink (root) |
Exposes .claude/agents/ (subagent definitions) to the plugin's default agent scan |
.devcontainer/ |
The repo's own dev container: a proven instance of templates/devcontainer.json |
- Copilot works through AGENTS.md, not
.claude/rules/*.md. - This file governs maintaining the repo.
- To consume the handbook — set up a VPS, Codespace, dev machine, or project — start at guides/bootstrap.md.
Working rules
- Read the target directory — learn existing content and style before editing or creating a file.
- Verify before claiming — search the codebase before asserting anything about existing code, structure, or behaviour. Read the actual source instead of guessing.
- Decide before you ask — a question is the last resort, not the opening move. Full gate: .claude/skills/clarify/question-rules.md.
- Web search for external knowledge — when working with external tools, libraries, or specs, consult authoritative sources. Use official docs and RFCs, not training data.
- Single source of truth — never duplicate content across files. Reference a template, script, or another doc with a relative link instead of copying it inline.
- Read
README.mdfirst; update it after every add, remove or rename. - No dead links — after renaming or deleting a file,
grep -r '<filename>' .and update or remove every reference.make linkschecks file targets only. - Current state only — docs, comments and instructions describe what is true now.
- Git history is the archive, and the only record of a prior state.
- A change that makes a statement false rewrites or deletes it in the same change.
- A superseded version never stands beside its replacement.
- Banned in prose: dated change entries, "previously / formerly / used to", deprecation notes.
- Exceptions:
CHANGELOG.md, ADR files, git history. - Delete, don't deprecate — a redundant file is deleted, with every reference removed.
- Version consistency — when a tool version changes,
grepthe whole repo and update every occurrence. - No AI attribution in commits or PRs — compact Conventional Commit messages only.
- Never append
Co-Authored-By: Claude …,Claude-Session: …,🤖 Generated with …, or similar trailers/footers. - The ban holds even when the session harness instructs it by default.
- The cloud-session GitHub tooling silently injects a
_Generated by Claude Code…_trailer into PR bodies on create. - It injects even when the submitted body has none.
- After creating a PR, re-read its body and strip the trailer with a body update.
- Body updates are not re-injected.
- Never append
- PR titles use Conventional Commit format (like commit messages). A squash merge of a PR with 2+ commits turns the PR title into the commit message. A single-commit PR keeps its own commit title.
- Ask before deleting or renaming a file (check for references first).
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.
- today Changed · -3 lines · +18 tokens per session c853f112e718
- 3d ago First seen · 115 lines · 1,514 tokens per session scan A bbe1014208bf
handbook AGENTS.md is an instructions file published in the GitHub repository nicograef/handbook (2 stars, last pushed today), licensed MIT. It adds 1,532 tokens to every session, about $0.0077 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.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.