Borrowing it
Nothing to install: this file belongs to stanveer/Biomedical-Agent-DB-Karpathy. 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/stanveer/Biomedical-Agent-DB-Karpathy/master/CLAUDE.mdgit clone --depth 1 https://github.com/stanveer/Biomedical-Agent-DB-KarpathyWrote 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/stanveer/biomedical-agent-db-karpathy/claude-md)<a href="https://agentmods.dev/instructions/stanveer/biomedical-agent-db-karpathy/claude-md"><img src="https://agentmods.dev/badge/instructions/stanveer/biomedical-agent-db-karpathy/claude-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/stanveer/biomedical-agent-db-karpathy/claude-md"><img src="https://agentmods.dev/badge/instructions/stanveer/biomedical-agent-db-karpathy/claude-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 | $0.00553 | $0.00553 |
| Opus 5 | $0.00277 | $0.00277 |
| Sonnet 5 | $0.00111 | $0.00111 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
Biomedical-Agent-DB-Karpathy CLAUDE.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 9d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Biomedical database guidelines for coding agents. Merge with project-specific instructions as needed.
Tradeoff: These instructions bias toward correctness, provenance, and cautious normalization over speed.
1. Normalize, Don't Guess
Do not invent identifiers, ontology mappings, or biological interpretations.
Before implementing or transforming data:
- State which namespaces are considered canonical for the current task.
- If a symbol or term is ambiguous, preserve ambiguity instead of collapsing it silently.
- If the source does not provide enough context, leave the field unresolved and record why.
- Distinguish normalization from enrichment. Do not add extra claims just because they seem biologically plausible.
2. Keep Entity Types Separate
Genes, proteins, variants, diseases, chemicals, pathways, and papers are different objects.
- Never treat a gene symbol as a protein accession.
- Never store a disease label as if it were a free synonym for a phenotype, condition, or indication without source support.
- Never collapse a variant string into a single canonical form without genome build and representation context.
- Preserve species explicitly whenever the source can refer to more than one organism.
3. Provenance First
Every nontrivial claim should be traceable to a source.
- Store source namespace, source record ID, release version, and ingest timestamp.
- Keep extracted evidence separate from curated assertions.
- When two sources disagree, preserve both assertions and their provenance rather than overwriting one with the other.
- Do not summarize away uncertainty that the source expresses.
4. Minimal, Verifiable Schema Changes
Touch the schema only when needed, and explain why.
- Prefer extending ingestion code over changing core tables unless the current schema blocks a required behavior.
- If a schema change is needed, describe the exact behavioral gap first.
- Do not add speculative columns for future use.
- Any new field should answer a current ambiguity, provenance need, or query need.
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.
- 9d ago First seen · 64 lines · 553 tokens per session scan A de1a55d38cf4
Biomedical-Agent-DB-Karpathy CLAUDE.md is an instructions file published in the GitHub repository stanveer/Biomedical-Agent-DB-Karpathy (2 stars, last pushed 4mo ago), licensed MIT. It adds 553 tokens to every session, about $0.0028 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
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.
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.