ultracite is a code linter and formatter that applies a fixed set of style and quality rules without requiring project-specific configuration. It is for developers who want automated checks and formatting for their code.
Borrowing it
Nothing to install: this file belongs to haydenbleasel/ultracite. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/haydenbleasel/ultracite/main/AGENTS.mdgit clone --depth 1 https://github.com/haydenbleasel/ultraciteWrote 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/haydenbleasel/ultracite/agents-md)<a href="https://agentmods.dev/instructions/haydenbleasel/ultracite/agents-md"><img src="https://agentmods.dev/badge/instructions/haydenbleasel/ultracite/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/haydenbleasel/ultracite/agents-md"><img src="https://agentmods.dev/badge/instructions/haydenbleasel/ultracite/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.05864 | $0.05864 |
| Opus 5.5 | $0.02346 | $0.02346 |
| Sonnet 5.5 | $0.01173 | $0.01173 |
| Haiku 4.5 | $0.00586 | $0.00586 |
Grade C, and why
ultracite AGENTS.md scanned grade C with 2 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 3d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
`bun run bump-tools` updates the linter/formatter packages; `bun run bump-deps` updates everything else. Upgrading oxfmt has needed a clean reinstall (`rm -rf node_modules && bun install`) and a regenerated `patches/oxfm Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- Spawn processes through `spawn-sync.ts`, never `node:child_process` directly. It normalises Windows `.cmd` shims and is what tests mock. How it starts
The opening of the file, as written. The whole thing — 207 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guide for coding agents working in this repository. The generated code standards that Ultracite itself enforces live in .claude/CLAUDE.md; this file covers how the repo is put together and how to change it safely.
What this is
Ultracite is a zero-config linting and formatting preset for JS/TS projects, published to npm as ultracite. It ships:
- Presets for three linter backends: Oxlint + Oxfmt (recommended), Biome, and ESLint + Prettier + Stylelint, each with core and per-framework variants.
- A CLI (
ultracite init | check | fix | doctor | upgrade) that installs the toolchain, writes config files, wires editors, git hooks, and AI agent rules, and shells out to the chosen linter. - An agent skill (
skills/ultracite) that ships inside the npm package.
The docs site at https://www.ultracite.ai lives in this repo too.
Repo map
packages/cli/ The published `ultracite` package
src/index.ts Commander entry; every subcommand is registered here
src/commands/ check, fix, doctor, upgrade
src/initialize.ts `ultracite init` orchestration (prompts + flags)
src/linters/ One adapter per tool: biome, eslint, oxlint, oxfmt, prettier, stylelint
src/integrations/ husky, lefthook, lint-staged, pre-commit
src/agent-fix/ `fix --claude` / `fix --codex`: hand remaining diagnostics to an agent CLI
src/data/ Static tables: agents, editors, hooks, options, providers, rules (AGENTS.md template)
src/dependencies.ts Toolchain versions + peer ranges read from package.json; what `init` installs
config/ The presets (see "Presets" below)
__tests__/ bun:test suite, one file per source module, plus lint-for-real fixtures
scripts/ generate-dts, copy-skill, compare-rule-parity, vendor-anti-slop
build.ts Bun.build -> dist/index.js (single minified ESM file, deps external)
apps/docs/ Docs + marketing site (blume on Astro, deployed to Cloudflare Workers)
docs/**/*.mdx Documentation content; meta.ts files order the sidebar
blog/*.mdx Research posts (served at /research), each with a page in pages/research (see "Research posts" below)
pages/ Custom .astro routes: the homepage, landing pages, the research index and posts
blume.config.ts Site config, content sources, redirects under /docs/*
packages/typescript-config/ Shared tsconfig bases (private)
skills/ultracite/ Source of the agent skill; copied into packages/cli/skills on pack
benchmark/ PR-time performance regression gate for check/fix across all providers
research/ Research harnesses: verified bibliography, real-codebase corpus, agent fix experiment (see research/README.md)
scripts/ validate-configs (loads every preset + runs the ESLint/oxlint parity check)
patches/ bun patchedDependencies (currently oxfmt)
.changeset/ Changesets; every user-facing change needs one
tmp/ Gitignored scratch area for hand-testing the CLI against a sample project
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.
- 3d ago Changed · -2 lines · +58 tokens per session cb0ca842b8ab
- 7d ago Changed · -1 lines · -31 tokens per session d23a1757bd8a
- 10d ago Changed · +31 lines · +1,890 tokens per session 661d370841c9
- 15d ago Changed · +55 tokens per session 79476f21c635
- 1mo ago First seen · 179 lines · 3,892 tokens per session scan C b1f085cc9c0d
ultracite AGENTS.md is an instructions file published in the GitHub repository haydenbleasel/ultracite (3,348 stars, last pushed today), licensed MIT. It adds 5,864 tokens to every session, about $0.0235 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-20.
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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 copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.