Claude Plugin
Last commit on June 23, 2026·Created on March 13, 2026
uditgoenka/autoresearch
“Routes a plain-language goal through debug, fix, audit, and ship until the predicate is satisfied.”
Combined rank
#54
across all skills
In Research
#4
category rank
Stars
5.8k
+7.0% in last 7d
Forks
439
+9.2% in last 7d
Watchers
23
+4.5% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of real attention, adoption, and active maintenance.
Description
This skill turns a plain-language goal into an autonomous pipeline. It classifies the goal into an archetype, then routes through subcommands like debug, fix, security, and ship in a bounded loop until a success predicate is met. The orchestrator enforces safety invariants: no auto-deploy, bounded iterations, and logged state. It helps AI agent power users, platform teams, software engineers, and security engineers automate iterative workflows with guardrails.
WHO IT'S FOR
AI Agent Power Users
automate iterative code improvement loops
Platform / DevEx Teams
standardize AI-driven development workflows
AI-Assisted Software Engineers
debug and fix code autonomously
Security Engineers
conduct automated security audits
30 Days of GitHub Stars
+517 stars↑ 9.9% monthly change
Jul 7Jul 14Jul 21Jul 28Aug 5
Compatibility & Install5 Skills · 3 Compatible agents · 2 categories · 5 install paths
Compatible agents
❯Claude Code❯Codex❯OpenCode
Categories
ResearchAutomation
Install viaRepository instructions Repository instructions Repository instructions Repository instructions
Repository instructions
Clone to Claude
Clone to Codex
Manual Copy
NPX Skills Add
Also in this category4 of 29