By DaisyCatTs
Autonomous research loop inspired by karpathy/autoresearch — give it a plain-language goal and it infers the artifact, evaluator (numeric scorer / objective gate / LLM rubric), and bounds, then iterates: cheap sequential rounds that escalate to a parallel GAN tournament when they plateau
Cancel active autoresearch loop
Explain autoresearch plugin and available commands
Autonomously turn a plain-language goal into a research loop — infers the artifact, evaluator, and bounds, then iterates (sequential, escalating to a tournament when stuck)
English | 简体中文
A curated collection of 15 plugins for Claude Code, providing specialized agents, skills, and automation tools for development and productivity workflows.
Conventional Git automation for commits and repository management with AI code quality checks.
Installation:
claude plugin install git@frad-dotclaude
GitFlow workflow automation for feature, hotfix, and release branches with semantic versioning.
Installation:
claude plugin install gitflow@frad-dotclaude
GitHub project operations with quality gates, TDD workflows, and comprehensive validation.
Installation:
claude plugin install github@frad-dotclaude
Advanced development workflow orchestration with BDD support and self-improving skills.
Installation:
claude plugin install superpowers@frad-dotclaude
Code simplification and refactoring with language-specific patterns and cross-file optimization.
Installation:
claude plugin install refactor@frad-dotclaude
SwiftUI Clean Architecture reviewer for iOS/macOS development with best practices enforcement.
Installation:
claude plugin install swiftui@frad-dotclaude
Generate a CLAUDE.md through an interactive workflow with environment detection, BDD/TDD testing options, and local best-practices references.
Installation:
claude plugin install claude-config@frad-dotclaude
Patent application generation, Feishu document creation, Product Requirements Documents, agent-based browser automation, Lark/Feishu CLI operations, and AI writing trope detection.
Installation:
claude plugin install office@frad-dotclaude
Validate and optimize Claude Code plugins against official best practices with agent-based fixes.
Installation:
claude plugin install plugin-optimizer@frad-dotclaude
Web frontend development toolkit — shadcn/ui, Next.js DevTools, React best practices, Supabase, DESIGN.md design system spec, and impeccable design skills.
Installation:
claude plugin install frontend@frad-dotclaude
Autonomous research loop inspired by karpathy/autoresearch — you supply an editable artifact, a scorer that prints one number, and an optimization direction; it runs bounded experiments, keeps a change only if the score improves, logs to results.tsv, and iterates via a stop hook. Works on any objective, not just ML training.
Installation:
claude plugin install autoresearch@frad-dotclaude
Delegate tasks and deep research to Google Gemini Managed Agents (Antigravity) running in a remote sandbox with code execution, Google Search, and URL reading, then read the results back. Runs asynchronously and polls for completion via the Monitor tool. Requires GEMINI_API_KEY and uv.
Installation:
claude plugin install antigravity@frad-dotclaude
Knowledge base for acpx - a headless ACP CLI for agent-to-agent communication.
Installation:
claude plugin install acpx@frad-dotclaude
Five methods to retrieve code context: DeepWiki, Context7, Exa, git clone, and web search.
Installation:
claude plugin install code-context@frad-dotclaude
General-purpose utility skills for documentation, writing, and project maintenance.
Installation:
claude plugin install utils@frad-dotclaude
plugin-name/..claude-plugin/plugin.json with required metadata..claude-plugin/marketplace.json./utils:update-readme to sync both README files.No model invocation
Executes directly as bash, bypassing the AI model
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Validates and optimizes Claude Code plugins against official best practices and file patterns
GitHub project operations with quality gates.
Agent and skills for code simplification and refactoring to improve code quality while preserving functionality
Hardware and EDA toolkit. Drives the KiCad CLI for schematic/PCB export and checks, and OpenSCAD for parametric 3D/2D part design and STL/DXF/PNG output.
Retrieve code context for any repo, library, or natural-language query via DeepWiki, Context7, Exa, git clone, and web search+fetch
npx claudepluginhub daisycatts/dotclaude --plugin autoresearchHarness-native ECC operator layer - 67 agents, 278 skills, 94 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
v9.52.0 - Reliability wave: tangle contextual review correction loop with hard round ceiling, progress-supervised review rounds (per-agent stall watch, descendant-tree kills), council diversity and agy pin fixes, marketplace generator source-of-truth fix, provider troubleshooting runbook and cost-expectations docs. Run /octo:setup.
Evidence-gated AI coding workflow: scan → analyze → plan → TDD → execute → fix → verify → review, powered by Codebase Memory MCP >= 0.9.0 with optional Serena LSP intelligence. Includes blast-radius planning, test/cycle gates, independent review, and Windows Git Bash hook auto-resolution.
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.