{
 "schema_version": 1,
 "generated": "2026-10-20",
 "id": "lobehub-awesome-rsi--pick-benchmarks-for-self-improvement-tests",
 "repo": "lobehub/awesome-rsi",
 "repo_url": "https://github.com/lobehub/awesome-rsi",
 "repo_slug": "lobehub-awesome-rsi",
 "title": "Pick benchmarks for testing a self-improving agent",
 "page_title": "How to pick benchmarks for testing a self-improving agent with Awesome Rsi",
 "summary": "Shortlist RSI-specific and proxy benchmarks from the list to test an agent you are building.",
 "who": "An AI engineer who is building a self-improving agent and needs an evaluation set before changing memory, prompts, or harness code.",
 "problem": "Generic agent benchmarks do not measure whether an agent improves its own improvement process, so the engineer needs benchmarks that match the RSI question.",
 "outcome": "A shortlist of three to five benchmarks, each labeled as direct RSI evaluation or as a proxy, with the reason for each choice.",
 "steps": [
  "Open the Benchmarks & Evaluations section and read its introduction.",
  "Review the Direct RSI & Self-Improvement Evaluations subsection first and note any entries that match your agent's update target.",
  "Review the AI R&D Capability Proxies and Agent Capability Proxies subsections for useful stand-in tests.",
  "Open the frontier lab evaluation frameworks subsection and note any framework that covers the risk you care about.",
  "Write the shortlist with one line per benchmark stating whether it is direct or a proxy and what it would check in your agent."
 ],
 "agent_prompt": "Using the Benchmarks & Evaluations section of the awesome-rsi README at lobehub/awesome-rsi, shortlist three to five benchmarks for an agent that updates its <update target>. Label each as direct RSI evaluation or capability proxy and give one reason per choice. Verify that every benchmark name and link is copied from the README.",
 "agents": [
  "any"
 ],
 "difficulty": "Intermediate",
 "time": "30 minutes",
 "basis": "readme",
 "caveat": "The list links to benchmarks but does not say how to run them, so setup for each benchmark must come from its own source.",
 "category": "Research",
 "tags": [
  "Research"
 ],
 "lang": "",
 "requirements": [
  "A web browser to open the GitHub README and the linked arXiv or journal pages",
  "Optional: a note-taking file or spreadsheet to track papers and their evidence levels"
 ],
 "risk": "none",
 "seen_in_the_wild": false,
 "mentioned_publicly": false,
 "risks": [],
 "best": false,
 "score": 0.8,
 "evidence": [],
 "url": "https://dedupedai.com/use-cases/lobehub-awesome-rsi/pick-benchmarks-for-self-improvement-tests/",
 "md": "https://dedupedai.com/use-cases/lobehub-awesome-rsi/pick-benchmarks-for-self-improvement-tests/index.md",
 "disclaimer": "Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked \"Seen in the wild\", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email hello@dedupedai.com and we will look at it promptly.",
 "api": "https://dedupedai.com/api/v1/usecases/lobehub-awesome-rsi--pick-benchmarks-for-self-improvement-tests.json",
 "api_version": "v1"
}