Borrowing it
Nothing to install: this file belongs to drivelineresearch/autoresearch-claude-code. 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/drivelineresearch/autoresearch-claude-code/master/AGENTS.mdgit clone --depth 1 https://github.com/drivelineresearch/autoresearch-claude-codeWrote 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/drivelineresearch/autoresearch-claude-code/agents-md)<a href="https://agentmods.dev/instructions/drivelineresearch/autoresearch-claude-code/agents-md"><img src="https://agentmods.dev/badge/instructions/drivelineresearch/autoresearch-claude-code/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/drivelineresearch/autoresearch-claude-code/agents-md"><img src="https://agentmods.dev/badge/instructions/drivelineresearch/autoresearch-claude-code/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 | $0.00418 | $0.00418 |
| Opus 5.5 | $0.00167 | $0.00167 |
| Sonnet 5.5 | $0.00084 | $0.00084 |
| Haiku 4.5 | $0.00042 | $0.00042 |
Grade A, and why
autoresearch-claude-code 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.
What it actually says
Contributor guidance
This repository supports a shared autoresearch skill for Claude Code and Codex. Reviewing or modifying this plugin does not authorize starting an experiment loop.
skills/autoresearch/SKILL.mddefines workflow and scope;references/state.mddocuments the schema. Update both when behavior changes.skills/autoresearch/scripts/ar_state.pyowns state validation, budgets, transactional logging, and Claude hook behavior. Shell hooks resolve their actual path (including manual-install symlinks) before dispatching here.skills/autoresearch/scripts/codex_loop.pysupervises already initialized experiments throughcodex exec. Keep continuation bounded; never enable sandbox/approval bypasses as an installation default.scripts/install.pyowns installation. Preserve foreign files, links, settings, and exact hook ownership. Test only with fixture homes, not real user config.- Experiment cleanup/staging must use explicit file scope. Do not introduce blanket git cleanup or overwrite experiment history on resume.
- Hook stdout must match the Claude event's output contract. Diagnostics go to stderr. Invalid state must not trap the agent in forced continuation.
- Use
uvfor example package management. Keep data/scoring/splits separate from candidate changes; no outer-fold labels in training, scaling, or selection. - Run
python3 -m unittest discover -s tests -vand ShellCheck after runtime changes. Scientific example tests may skip without deps; report that separately. - Preserve ignored
brain/, runtime state, and unrelated work. Avoid running experiments, making network/model calls, or modifying personal installs as a side effect of tests. Tests use temporary repositories and fake clients.
Layout: commands/ and hooks/hooks.json are Claude-specific adapters;
skills/autoresearch/ is a self-contained shared skill; examples/ is an optional
ML workload; docs/review.md records review evidence and remaining work.
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 · 31 lines · 418 tokens per session scan A 969f70407e37
autoresearch-claude-code AGENTS.md is an instructions file published in the GitHub repository drivelineresearch/autoresearch-claude-code (348 stars, last pushed 2mo ago), licensed MIT. It adds 418 tokens to every session, about $0.0017 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 2026-09-08.
autoresearch-claude-code AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| autoresearch-claude-code AGENTS.mddrivelineresearch/autoresearch-claude-code | Instructions file | 348 | 2mo ago | MIT | A | 418 | 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 | today | 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 |