Borrowing it
Nothing to install: this file belongs to LumeWeb/web. 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/LumeWeb/web/develop/AGENTS.mdgit clone --depth 1 https://github.com/LumeWeb/webWrote 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/lumeweb/web/agents-md)<a href="https://agentmods.dev/instructions/lumeweb/web/agents-md"><img src="https://agentmods.dev/badge/instructions/lumeweb/web/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/lumeweb/web/agents-md"><img src="https://agentmods.dev/badge/instructions/lumeweb/web/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 | $0.01421 | $0.01421 |
| Opus 5.5 | $0.00568 | $0.00568 |
| Sonnet 5.5 | $0.00284 | $0.00284 |
| Haiku 4.5 | $0.00142 | $0.00142 |
Grade A, and why
web 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 1mo 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Shared rules for every coding agent and PR reviewer in this repo, a pnpm
monorepo with React and TypeScript under apps/ and libs/, and Go services
under go/. The "Writing" section applies to every kind of prose an agent produces
here: code comments, commit bodies, PR descriptions, test names, and replies
to users.
Writing
Plain words. Name the real thing: the function, the file, the platform, the failure. Use the word you would say out loud, and prefer a verb to a noun built from one ("validates", not "performs validation of"). No abstract sentence where a fact belongs.
Accuracy. A sentence stating a checkable fact (a field list, a state set, a count, a returned shape, a path, a script name) must match the code. Verify it or leave it out; a reader trusts a comment over the code.
Numbers. A number the code or the contract fixes (a 200ms interval, a retry cap) passes in a comment, and so does one derived from the mechanism. Do not hedge them: a fixed 2x stays "2x", not "~2x". A number someone had to run something to get (a benchmark result, a profile percentage, a measured timing) fails in a comment; it drifts while still reading as fact. State the mechanism in the comment and put the measurement in the PR description, where its conditions live.
Never write: delve, leverage, utilize, robust, seamless, crucial, pivotal, paramount, realm, underscore, foster, elevate, unlock, streamline, boast, landscape, holistic, bespoke, groundbreaking, testament, synergy, endeavor, meticulous, nuanced, notably, importantly, essentially, effectively, simply, basically, additionally, furthermore, moreover, comprehensive, myriad, plethora, facilitate, elegant, load-bearing, slice, gate, seam, or "landed" for shipped or merged. No em-dashes, no arrows, no "not just X, it's Y" pivot, no negative-flip contrast, no rule of three, no over-bolding. Colons, semicolons, and parentheses carry those jobs.
A word on that list is fine where it is the codebase's own term: a function name, an API verb, a field, a character. The list targets filler.
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.
- 1mo ago First seen · 118 lines · 1,421 tokens per session scan A 4481c822450a
web AGENTS.md is an instructions file published in the GitHub repository LumeWeb/web (2 stars, last pushed yesterday), licensed MIT. It adds 1,421 tokens to every session, about $0.0057 per session on Opus 5.5. 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 2026-09-28.
web AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| web AGENTS.mdLumeWeb/web | Instructions file | 2 | yesterday | MIT | A | 1,421 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |