AIDE²-style recursive self-improvement loop: an outer agent iteratively rewrites an inner tree-search research agent, keeping a rewrite only when it beats the incumbent on private held-out scores under a fixed budget across an optimization / ML-engineering / harness-engineering battery. Standalone /rsi:autoresearch solves tasks by tree search; /rsi:init + /rsi:step + /rsi:run drive the outer loop.
Plugin rsi-loop from marketplace claude-skills in cajias/claude-skills. Commands read from the repo's manifest — 1 month ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
autoresearchStandalone AIDE-style autoresearch agent: tree-search over candidate solutions (draft/debug/improve operators) for any task directory with a task.md and score.py — ML engineering, heuristic optimization, or harness engineering. Runs the current best generation evolved by the rsi-loop outer loop (fal
rsi-loopAIDE²-style recursive self-improvement outer loop: propose a rewrite of the inner research agent, evaluate it on a heterogeneous task battery under a fixed budget, and keep it only if it beats the incumbent on private held-out scores. Use when the user wants to run or understand an rsi-loop outer st
/rsi-loop:rsi-autoresearchRun the inner tree-search autoresearch agent standalone on a task directory
/rsi-loop:rsi-igniteLevel-2 ignition test — does a campaign driven by the best evolved generation beat the baseline campaign at equal budget?
/rsi-loop:rsi-initScaffold an rsi-loop run directory (generations, ledger, best pointer)
/rsi-loop:rsi-reportCompute RSI-ladder evidence for a run — improvement slope, human-baseline delta, holdout generalization, hack-rate trend
/rsi-loop:rsi-runDrive the outer loop for many steps (propose→evaluate→verify→select) until a stop condition
/rsi-loop:rsi-stepRun one AIDE²-style outer-loop step (propose → evaluate → verify → select)
rsi-proposerOuter-loop proposer for the rsi-loop plugin. Writes the next inner-agent generation as a mutation of the incumbent best, driven by the run ledger. Use only from /rsi:step.
rsi-verifierAdversarial verifier for rsi-loop results. Audits a candidate generation's winning solutions for reward hacking before selection. Use only from /rsi:step.
sh ${CLAUDE_PLUGIN_ROOT}/hooks/deny-private-hook.sh
Hard-won gotchas from authoring and distributing Claude Code plugins, skills, hooks, and Workflow scripts. Bundles APM packaging traps, plugin cache and marketplace path failures, hook main-vs-subagent detection, Workflow meta parsing breaks, and orchestration verification playbooks.
16 skills
Design-and-build process skills: layered design docs, C4 architecture artifacts, Diataxis documentation, README authoring, a milestone execution harness, milestone goal-prompt generation, positive-framing spec guidance, and a schema drift playbook.
8 skills
Operational gotchas and references for third-party integrations driven from Claude Code — Cloudflare Workers, Grafana, Notion, OpenPencil, Playwright, and MCP file transport.
7 skills
Hard-won gotchas from real git and gh workflows: PR create/merge quirks, worktree races, rebase in non-interactive shells, squash-merge diagnosis, and deadlock recovery via plumbing.
6 skills
Hard-won gotchas from CI pipelines and build environments: GitHub Actions, Copilot CLI headless runs, cron block generation, Docker on Apple Silicon, and TypeScript monorepo scaffolding.
5 skills
Hard-won gotchas from the Python toolchain: importlib class identity, pip/PyPI name collisions, mock.patch.object vs __slots__, ruff on notebooks, and uv in fresh worktrees.
5 skills
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Backend API design, GraphQL architecture, workflow orchestration with Temporal, and test-driven backend development
9 skills · 1 command · 8 agents
AI-supervised issue tracker for coding workflows
1 skill · 1 agent · hooks
MiniMax AI skills library for frontend, fullstack, Android, iOS, shader, GIF sticker, document, presentation, spreadsheet, and multimodal media workflows
18 skills
Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. This plugin provides configuration guidance (LLM commit messages, project hooks, worktree paths), automatic activity tracking (🤖/💬 indicators in `wt list` showing active Claude sessions), and a `/wt-diff` pane showing each worktree's changes.
2 skills · hooks
GSD Core is a meta-prompting, context engineering, and spec-driven development system for AI coding agents.
72 skills · 72 commands · 64 agents · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.