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/context4ai/agent-graph/agents-mdgit clone --depth 1 https://github.com/context4ai/agent-graphWhat 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 | $0.00442 | $0.00442 |
| Opus 5 | $0.00221 | $0.00221 |
| Sonnet 5 | $0.00088 | $0.00088 |
| Haiku 4.5 | $0.00044 | $0.00044 |
Grade A, and why
agent-graph 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.
How it starts
The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Graph
README · Development · English manual · 中文手册
Agent Graph is an independent open-source project. It defines a Skills-native graph specification and ships a reference SDK and CLI for authoring, validating, testing, building, evaluating, and inspecting agent workflows.
Engineering rules
- Use Bun for development commands and tests.
- Keep all runtime code Node.js compatible. Bun-only APIs are allowed only in build and test tooling.
- Use ESM and strict TypeScript.
- Keep route consumption agent-led: evaluation and route resolution must not execute model calls or lifecycle Actions; explicit authoring, import, build, Run, and materialization commands may write only to caller-selected paths.
- Treat provider facts and content digests as authoritative. Run history is not proof that external work is complete.
- Do not hard-code Claude, Codex, npm-global, home-directory, or workspace paths into the protocol.
- Long instructions, schemas, templates, and context views remain addressable files; do not inline them into routine CLI output.
- New protocol fields require matching schemas, documentation, tests, and at least one example.
- Keep English and Simplified Chinese user documentation behaviorally equivalent.
- Generated output belongs in
dist/or.tmp/and must not be committed.
Commands
bun install
bun run typecheck
bun run lint
bun run test
bun run build
bun run verify
Documentation ownership
README.md/README.zh-CN.md: product entry and quick start.docs/en//docs/zh-CN/: user manuals and specification guides.case-studies/: Agent Graph-owned examples or compatibility redirects. Product-specific case content and replay data belong to the integrating product repository and are linked from here rather than duplicated.DEVELOPMENT.md: contributor workflow and release checks.- Schema files in
schemas/: machine-readable contracts; prose must not contradict them.
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.
- 2d ago First seen · 40 lines · 442 tokens per session scan A b065075b307d
agent-graph AGENTS.md is an instructions file published in the GitHub repository context4ai/agent-graph (12 stars, last pushed 2d ago), licensed MIT. It adds 442 tokens to every session, about $0.0022 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 2026-08-30.
Other instructions, from other repositories
output CLAUDE.md
Instructions for growthxai/output, covering claude.md, project overview, documentation references, tech stack context and environment variables.
vscode design-philosophy.instructions.md
VS Code design philosophy — the shared Values→Principles→Moves vocabulary for reasoning about UI in design terms rather than raw pixels. Use when creating, editing, or reviewing any visual surface (CSS, DOM, theming, icons, motion). Name the value/principle before reaching for a token.
deepagents AGENTS.md
Instructions for langchain-ai/deepagents, covering global development guidelines for the deep agents monorepo, corridor security analysis, development workflow, suppressing ruff rules and pr conventions.
workflow AGENTS.md
Instructions for vercel/workflow, covering agent instructions, architecture, core components, workflow execution model and development commands.
walkerOS CLAUDE.md
Instructions for elbwalker/walkerOS, covering claude.md and important: use agent.md.
huly-mcp CLAUDE.md
Instructions for dearlordylord/huly-mcp, covering project instructions, design principle: llm-first api, project harness (copy to new projects), package manager and verification.