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/oxc-project/oxc/agents-mdgit clone --depth 1 https://github.com/oxc-project/oxcWhat 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 | $0.03232 | $0.03232 |
| Opus 5 | $0.01616 | $0.01616 |
| Sonnet 5 | $0.00646 | $0.00646 |
| Haiku 4.5 | $0.00323 | $0.00323 |
Grade A, and why
oxc 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 — 330 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - AI Assistant Guide for Oxc
Oxc is a high-performance JavaScript/TypeScript toolchain written in Rust containing:
- Parser (JS/TS with AST), Linter (oxlint), Formatter (oxfmt), Transformer, Minifier
AI Usage Policy for Contributors
IMPORTANT: If you are an AI assistant helping a human contributor:
- Disclose AI usage - Contributors must disclose when AI tools were used to reduce maintainer fatigue
- Full responsibility - The human contributor is responsible for all AI-generated issues or PRs they submit
- Quality standards - Low-quality or unreviewed AI content will be closed immediately
All AI-generated code must be thoroughly reviewed, tested, and understood by the contributor before submission. Code should meet Oxc's performance and quality standards.
- Ban policy - Contributors who submit repeated low-quality ("slop") PRs will be banned without prior warning. Bans may be lifted if you commit to contributing to Oxc in accordance with the policy above. You may request an unban via our Discord.
Repository Structure
Rust workspace with key directories:
crates/- Core functionality (start here when exploring)apps/- Application binaries (oxlint, oxfmt)napi/- Node.js bindingsnpm/- npm packagestasks/- Development tools/automation
Avoid editing generated subdirectories.
Core Crates
oxc_parser- JS/TS parseroxc_ast- AST definitions/utilitiesoxc_semantic- Semantic analysis/symbols/scopesoxc_linter- Linting engine/rulesoxc_formatter- Code formatting (Prettier-like for JS)oxc_formatter_*- Code formatting (Prettier-like for other languages)oxc_transformer- Code transformation (Babel-like)oxc_minifier- Code minificationoxc_codegen- Code generationoxc_isolated_declarations- TypeScript declaration generationoxc_diagnostics- Error reportingoxc_traverse- AST traversal utilitiesoxc_allocator- Memory managementoxc_language_server- LSP server for editor integrationoxc- Main crate
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 · 330 lines · 3,232 tokens per session scan A 7e8dd476b66e
oxc AGENTS.md is an instructions file published in the GitHub repository oxc-project/oxc (22,563 stars, last pushed 2d ago), licensed MIT. It adds 3,232 tokens to every session, about $0.0162 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
js2 AGENTS.md
Instructions for loopdive/js2, covering agent instructions, claude memories, commit messages, commit attribution and codex session.
js2 CLAUDE.md
Instructions for loopdive/js2, covering js2wasm, answering style, running tests, dev scratch and working in worktrees.
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.
page-agent AGENTS.md
AGENTS.md instructions for alibaba/page-agent, covering instructions for coding assistants, project overview, development commands, architecture and monorepo structure.
biome AGENTS.md
Instructions for biomejs/biome, covering agent guidelines for contributing to biome, communication, before editing, implementation gate and final review.
ax AGENTS.md
AGENTS.md instructions for ax-llm/ax, covering ax repository guide, what this file is for, repository layout, canonical docs and current project defaults.