Skip to main content
skill CX Codex

autoresearch · pedronauck — Claude Code skill: what it does, how to install

autoresearch runs another skill through repeated test cycles, grades each output against pass/fail checks, and rewrites its prompt until the score stops climbing.

no license 152 installs ★ 655 updated 2026-10-20

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add pedronauck/skills --skill autoresearch -a claude-code
Open on GitHub

What autoresearch does

The skill takes the path to a target SKILL.md, a handful of test prompts, and 3–6 yes/no checks, then loops: generate outputs, grade them against the checks, tweak the wording, keep edits that raise the pass rate and drop the rest. Before any run starts, the agent has to confirm with the user which skill to tune, what test scenarios to cover, what the pass/fail checks are, how many runs per experiment (5 by default), and how often cycles repeat (every 2 minutes by default).

Before touching anything, the agent reads the full SKILL.md and any linked files under references/ to understand the skill's job, then turns the user's criteria into checks written as a question plus explicit pass and fail conditions. The first pass runs the skill unchanged as experiment zero, its score becomes the baseline, and the original file is copied to SKILL.md.baseline.

Every experiment gets logged to results.tsv and results.json, feeding a self-contained HTML dashboard with a score-progression chart, a per-experiment table, and a breakdown of which checks pass or fail most often; it auto-opens in the browser and refreshes as new runs land. If the baseline already lands near 90%, the skill flags that to the user before continuing.

What it helps with

  • find which inputs break an existing skill
  • turn vague quality bars into testable pass/fail checks
  • auto-tune a skill's prompt instead of editing it by hand
  • track pass-rate progress across experiments on a chart
  • log which prompt edits helped and which got discarded

Example prompts for the agent

  • Run autoresearch on the skill at ~/.claude/skills/pdf-report/SKILL.md
  • Benchmark this skill with evals until the pass rate plateaus
  • Improve this skill — set up 5 test scenarios and binary checks

How to install autoresearch

1Skills CLI (Vercel)

npx skills add pedronauck/skills --skill autoresearch -a claude-code

2Manually: put the files in place

~/.claude/skills/autoresearch/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/autoresearch/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about autoresearch

How do I install autoresearch in Claude Code?

Run npx skills add pedronauck/skills --skill autoresearch -a claude-code; the skill lands at ~/.claude/skills/autoresearch/SKILL.md.

What does autoresearch do?

It runs a target skill against test inputs, grades every output with binary pass/fail checks, and mutates the skill's prompt automatically, keeping only the versions that raise the pass rate.

What do I need before running autoresearch?

The exact path to the target skill's SKILL.md, 3 to 5 varied test scenarios, and 3 to 6 binary quality checks — the agent won't start experiments without these.

Similar skills

All skills for Agent setup: memory, context, orchestration →

autoresearch by other authors comes first.

autoresearch official luiscantero autoresearch — a GitHub Copilot skill that runs an autonomous experiment loop: it edits code, tests, measures a metric, and keeps only what improves it. The project must be a git repository with terminal access. CX Codex +1 3,400installs autoresearch uditgoenka The autoresearch skill changes code, checks it against a metric, then keeps or discards the result. Separate commands cover debugging, security audits, and other workflows. CX Codex +1 2,700installs autoresearch Yeachan-Heo autoresearch runs one mission through repeated change-and-evaluate cycles, writing each result to JSON and a markdown decision log. Useful when the work needs a long recorded search rather than one attempt. CX Codex +1 654installs autoresearch Orchestra Research autoresearch manages autonomous research projects using a two-loop cycle for rapid experiment execution and periodic synthesis. CX Codex +1 1,200installs autoresearch AgriciDaniel autoresearch runs a bounded research loop with web access, drafts a cited report, and lets you review before filing results. Use when you need deep investigation with fact-checking built in. CX Codex +1 572installs skill-autoresearch akillness skill-autoresearch runs edits to skills, prompts and SOPs through a frozen local benchmark: one change per run, then keep or revert. CX Codex +1 323installs skill-optimizer mcollina skill-optimizer is a Claude Code skill for refining other skills and testing how models follow their instructions. Use it while editing skill packs, investigating regressions, or setting release checks. CX Codex +1 1,400installs prompt-optimizer haabe prompt-optimizer — a Claude Code skill that A/B tests CLAUDE.md edits against eval benchmarks. Run it when agent instructions stop producing stable output and you need to measure which wording works. CX Codex +1 ★ 54repo stars write-judge-prompt ai-evals-course write-judge-prompt walks you through one binary Pass/Fail LLM-as-Judge evaluator for a subjective failure mode: tone, faithfulness, relevance. Useful when code checks fall short. CX Codex +1 1,300installs prompt-engineer-toolkit borghei The prompt-engineer-toolkit Claude Code skill tests production prompts and scores model outputs. Use it to compare revisions, investigate regressions, and gate prompt changes before release. CX Codex +1 106installs prompt-governance Alireza Rezvani prompt-governance manages production prompts as versioned code, builds automated eval pipelines, and prevents quality regressions across AI deployments. CX Codex +1 40installs

From the same set52

The whole set →
ship-pr ship-pr delivers finished work with documentation audits, release note grouping, compliant commits, and a pull request. CX Codex +1 181installs kb-yt-channel kb-yt-channel — a skill that builds Karpathy KB topics from YouTube channels: the agent fetches transcripts, organizes them into topic folders, validates structure, and creates search indexes. CX Codex +1 175installs spec-peer-review spec-peer-review runs one independent review of an approved spec, RFC, design document, or detailed PRD. It returns findings for the user to select. CX Codex +1 161installs yc-apply yc-apply is a skill for preparing and checking Y Combinator applications, founder-video notes, interview practice, and reapplication materials based on verified facts. CX Codex +1 158installs insta-master insta-master is a Claude Code skill for planning Instagram content, refining profiles, diagnosing reach, and designing growth or monetization experiments. CX Codex +1 131installs yt-master yt-master — a skill for YouTube work: sharpens the idea, title, thumbnail concept and script, then reads the metrics when views or CTR slip. CX Codex +1 128installs golang-master golang-master — a skill for Claude Code that guides Go implementation and review. It applies to work on errors, concurrency, tests, performance, and module layout. CX Codex +1 115installs vc-pitch-deck vc-pitch-deck is a Claude Code skill that creates, reviews, and revises investor decks and slides. It prepares materials for email, presentations, and demo days. CX Codex +1 102installs

Project sponsors

Companies that keep this analytics open