{
 "schema_version": 1,
 "generated": "2026-10-20",
 "repo": "Danielxu0208/rsi_eval",
 "name": "rsi_eval",
 "owner": "Danielxu0208",
 "slug": "danielxu0208-rsi-eval",
 "github": "https://github.com/Danielxu0208/rsi_eval",
 "what": "A lightweight Python SDK and CLI for evaluating recursive self-improvement methods across agent benchmarks.",
 "why": "",
 "language": "Python",
 "requirements": [
  "Python 3.12 or newer and git",
  "Docker Engine with Compose and Buildx, or Docker Desktop, for benchmark runs",
  "Codex CLI on PATH and access to a model for candidate runs",
  "Harbor 0.18.0 installed through the harbor extra for benchmark runs"
 ],
 "audience_note": "Developers and AI researchers who want to test agent candidates or self-improvement loops on existing benchmarks without writing their own execution and result handling.",
 "risk": "none",
 "section": "earlier",
 "weekly_rank": null,
 "category": "Software and DevOps",
 "url": "https://dedupedai.com/use-cases/danielxu0208-rsi-eval/",
 "use_case_count": 4,
 "use_cases": [
  {
   "id": "danielxu0208-rsi-eval--preview-benchmark-tasks-before-running",
   "title": "Preview benchmark tasks before spending model quota",
   "best": true,
   "url": "https://dedupedai.com/use-cases/danielxu0208-rsi-eval/preview-benchmark-tasks-before-running/",
   "json": "https://dedupedai.com/use-cases/u/danielxu0208-rsi-eval/preview-benchmark-tasks-before-running.json"
  },
  {
   "id": "danielxu0208-rsi-eval--run-codex-candidate-on-terminal-bench-task",
   "title": "Run a Codex candidate on one Terminal-Bench task",
   "best": false,
   "url": "https://dedupedai.com/use-cases/danielxu0208-rsi-eval/run-codex-candidate-on-terminal-bench-task/",
   "json": "https://dedupedai.com/use-cases/u/danielxu0208-rsi-eval/run-codex-candidate-on-terminal-bench-task.json"
  },
  {
   "id": "danielxu0208-rsi-eval--build-custom-rsi-loop-with-evaluate",
   "title": "Build your own self-improvement loop around evaluate()",
   "best": false,
   "url": "https://dedupedai.com/use-cases/danielxu0208-rsi-eval/build-custom-rsi-loop-with-evaluate/",
   "json": "https://dedupedai.com/use-cases/u/danielxu0208-rsi-eval/build-custom-rsi-loop-with-evaluate.json"
  },
  {
   "id": "danielxu0208-rsi-eval--run-bundled-rsi-method-on-experiment",
   "title": "Run a bundled RSI method on a prepared experiment",
   "best": false,
   "url": "https://dedupedai.com/use-cases/danielxu0208-rsi-eval/run-bundled-rsi-method-on-experiment/",
   "json": "https://dedupedai.com/use-cases/u/danielxu0208-rsi-eval/run-bundled-rsi-method-on-experiment.json"
  }
 ],
 "disclaimer": "Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk."
}