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/arnosdeus/agent-toolkit/agents-mdgit clone --depth 1 https://github.com/Arnosdeus/agent-toolkitWhat 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 | $0.00543 | $0.00543 |
| Opus 5 | $0.00271 | $0.00271 |
| Sonnet 5 | $0.00109 | $0.00109 |
| Haiku 4.5 | $0.00054 | $0.00054 |
Grade A, and why
agent-toolkit 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shared Agent Instructions
This file is the portable source of truth for Codex, Cline, and compatible agents. Project-specific instructions and existing design systems take priority.
Repository inspection
Never start coding blindly. Inspect package manifests, project structure, existing components, configuration, tests, documentation, design systems, and all applicable agent instructions. Use the repository's established commands.
Existing architecture
Extend the existing architecture and components whenever practical. Avoid introducing frameworks, dependencies, or abstractions without justification. Keep changes scoped to the request and do not modify unrelated files.
Frontend and UI quality
Follow the project's design system. Use semantic HTML, accessible names, keyboard navigation, visible focus, responsive layouts, and reusable components. Design loading, empty, error, sparse, and dense states. For significant UI work: apply the Vercel Web Interface Guidelines, review the design, verify meaningful flows with Playwright, fix findings, and verify again.
Use image-to-code for screenshot/reference-driven work and preserve working
architecture while iterating toward visual fidelity.
Testing
Do not claim a feature works without appropriate verification. Run focused tests, lint/type checks when available, and Playwright for meaningful frontend changes. Report commands and results honestly.
Security
Never commit secrets. Do not add unsafe shell execution, unnecessary network calls, insecure authentication changes, disabled security controls, or unjustified dependency changes.
Git and safety
Do not reset user work, delete unrelated changes, rewrite history, force-push, or modify production repositories while validating this toolkit unless explicitly asked. Review the diff and check for regressions before handoff.
Standard workflow
UNDERSTAND → INSPECT → PLAN → IMPLEMENT → TEST → REVIEW → VERIFY → REPORT
Use relevant skills rather than blindly executing every skill on every task.
Use task-skill for implementation, git-review before handoff, and focused
review skills when the change warrants 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 · 76 lines · 543 tokens per session scan A db6f0e39bb00
agent-toolkit AGENTS.md is an instructions file published in the GitHub repository Arnosdeus/agent-toolkit (2 stars, last pushed 9d ago), licensed MIT. It adds 543 tokens to every session, about $0.0027 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
vibe-coding-repository-standard AGENTS.md
Instructions for cz1993/vibe-coding-repository-standard, covering vcrs repository instructions, purpose, communication, change discipline and privacy and safety.
matcha AGENTS.md
AGENTS.md instructions for plumpslabs/matcha, covering 🍵 matcha — core ai directives, the 6-checkpoint filter, core principles, intensity levels and commands.
map-framework CLAUDE.md
Instructions for azalio/map-framework, covering map framework (mapify-cli) — agent instructions, what this repo is, critical invariant: template single-source render, skill catalog invariant and how to work in this repo.
cctop AGENTS.md
Instructions for st0012/cctop, covering agents.md - development guide for cctop, scope and sources, required development rules, driver workflow and github and pr rules.
pixir CLAUDE.md
Instructions for Ranvier-Technologies/pixir, covering pixir harness - legacy agent guide, current map and commands.
claude-cost-optimizer CLAUDE.md
Instructions for Sagargupta16/claude-cost-optimizer, covering claude.md - claude-cost-optimizer, purpose, file structure, writing conventions and python tools.