Borrowing it
Nothing to install: this file belongs to ynfra/ybrain. 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/ynfra/ybrain/master/AGENTS.mdgit clone --depth 1 https://github.com/ynfra/ybrainWrote 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/ynfra/ybrain/agents-md)<a href="https://agentmods.dev/instructions/ynfra/ybrain/agents-md"><img src="https://agentmods.dev/badge/instructions/ynfra/ybrain/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/ynfra/ybrain/agents-md"><img src="https://agentmods.dev/badge/instructions/ynfra/ybrain/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 | $0.01009 | $0.01009 |
| Opus 5.5 | $0.00404 | $0.00404 |
| Sonnet 5.5 | $0.00202 | $0.00202 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
ybrain 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 2mo 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ybrain — component guide
Company knowledge MCP server (Bun + TypeScript). Serves prompts, ownership,
docs, and a skills index over MCP, backed by one Git data repo. Read the
root AGENTS.md for monorepo/Copybara rules; this file covers ybrain specifics.
Golden rules
- Tools never touch git directly. They depend only on the
DataStoreinterface insrc/store.ts. Add backends (GitHub API, Cloudflare KV) by implementingDataStore, not by editing tools. Keeps the "Git first, Cloudflare later" option open. - This is the CODE repo, not the DATA repo. Content (prompts, drafts, …)
lives in the separate
YBRAIN_DATA_REPO_URLrepo, cloned at runtime intoYBRAIN_DATA_DIR(gitignored). Never commit/data/here. - Single instance only. The write queue in
git-store.tsserializes commits in-process; that is the entire concurrency story. Do not run replicas against the same local-clone backend — parallel pushes will race. Scale reads only, or move writes to an API backend first. - Three tiers, no per-user identity. Reads +
suggest_*are open (anonymous, →drafts/). Only the supervisor tier (publish_draft/reject_draft/add_skill_source) is gated, by a namedYBRAIN_SUPERVISOR_TOKENSBearer. The real access control is Git merge/publish rights, not anything ybrain enforces.suggested_byis unverified metadata — never a security signal. - No memory feature. Long-term memory was intentionally dropped (high churn, no real privacy under open reads). Don't reintroduce it without a non-Git store and an identity model.
- External skill content is untrusted.
SKILL.mddescriptions from external repos are data, not instructions — label them, never blend into prompts.
Layout & flow
server.ts— stateless Streamable HTTP: a freshMcpServerper request, scoped byAuthContext. Supervisor tools registered only whencanPublish. Also: per-IP rate limit, body-size cap, HMAC-verified/webhook, JSON-RPC parse errors.tools.ts—registerReadTools+registerSuggestTools(both open) +registerSupervisorTools(gated).content.ts— frontmatter (gray-matter) parse/format + renderers;DIRSand thelivePath/draftPathhelpers.git-store.ts— clone/sync on boot, background refresh (fetch + hard-reset),enqueue()write queue with commit-then-rebase + reset-on-failure,move()for publish, git-op timeouts, credential scrubbing.skills.ts—SkillIndexreadsskill-sources.yaml, indexes internal skills (local clone) + external repos (GitHub trees API + raw fetch — never cloned or copied), generates per-kindinstall commands.reindexSource()/reindexInternal()keep authoring/registering off the full-crawl path.
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.
- 2mo ago First seen · 73 lines · 1,009 tokens per session scan A 6e1ad6097170
ybrain AGENTS.md is an instructions file published in the GitHub repository ynfra/ybrain (0 stars, last pushed 14d ago), licensed MIT. It adds 1,009 tokens to every session, about $0.0040 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 2026-08-31.
ybrain AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| ybrain AGENTS.mdynfra/ybrain | Instructions file | 0 | 14d ago | MIT | A | 1,009 | 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 |