claude

self-improvement-loop

Background reflection on Claude Code sessions that promotes recurring lessons into skills, hooks, rules and agents, with a local web UI to review and rate them

0 on GitHubby Mariusz Rakusv1.6.1MITupdated 2 days agoproductivity
/

Plugin self-improvement-loop from marketplace kolezka in kolezka/marketplace. Commands read from the repo's manifest — 2 days 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.

What it installs

Skills (2)

  • self-improvement-loop

    Use when working with the self-improvement loop from OpenClaw - reading the lesson block in the workspace file, queueing this session for reflection, listing reflections or lessons, or rating a staged artifact with good/bad feedback.

  • self-improvement-loop

    Use when operating the self-improvement loop (`sil`) from Claude Code or Codex, such as checking whether the worker runs, reading reflections or pending lessons, reviewing, accepting, rejecting or retiring a staged skill, hook, rule or agent, rating an artifact good or bad, or a ModelNotConfigured e

Slash commands (6)

  • /self-improvement-loop:curriculum

    Preview or run curriculum promotion for the current world

  • /self-improvement-loop:feedback

    Record a good or bad vote on an artifact

  • /self-improvement-loop:learn

    Turn a mistake caught in this session into a rule, skill, agent or nudge

  • /self-improvement-loop:loop

    Show loop status, or start the web UI / trigger a worker pass

  • /self-improvement-loop:reflect

    Queue a background reflection on the work just completed

  • /self-improvement-loop:toolify

    Turn the last working answer in this session into a reusable skill

Hooks (7)

  • SessionStart

    [ "$SIL_HOOK_MODULE" = "$PPID" ] || sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

  • UserPromptSubmit

    [ "$SIL_HOOK_MODULE" = "$PPID" ] || sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

  • PreToolUse

    [ "$SIL_HOOK_MODULE" = "$PPID" ] || sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

  • PostToolUse

    [ "$SIL_HOOK_MODULE" = "$PPID" ] || sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

  • Stop

    sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

  • SubagentStop

    sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

  • SessionEnd

    sh "${CLAUDE_PLUGIN_ROOT}/scripts/hook"

View source on GitHub

More from kolezka/marketplace

All 12 plugins →

Durable work-state persistence for Claude Code sessions, backed by the Outline MCP server. Reads prior context before work and records task start, checkpoints, handoff and completion

1 skill · 10 commands · 1 MCP server · hooks

0

Role subagents for orchestrated engineering work: explorer, codebase-researcher, architect, implementer, debugger, test-engineer, reviewer, security-reviewer and verifier

1 skill · 9 agents

0

Create and maintain evidence-based block documentation

1 skill · 2 agents

0

Install codegraph and graphify in a repository, route the agent to the right graph with near-zero hook overhead, and build the graphify graph on your own LLM backend (LiteLLM, Ollama, OpenAI-compatible) instead of agent subagents

2 skills · hooks

0

Web research through self-hosted SearXNG and Crawl4AI without MCP. Routes built-in web search and fetch through a local CLI. Needs Bun, Python 3.10+ and configured services. Private repository

1 skill · hooks

0

Design and redesign UI as single-file HTML drafts drawn by the current model, checked against real renders in light and dark, and compared side by side on a local canvas. No design service; needs Bun and Chrome. Private repository

1 skill

0

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

nextjs

claude

Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.

6 skills

vercel/next.js 143,495

caveman

claude

Talk like caveman. Cut filler. Keep technical details explicit.

22 skills · 1 command · 3 agents · hooks

JuliusBrussee/caveman 119,462

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Taste Skill

claudecodex

tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.

1 skill

Leonxlnx/taste-skill 114,391

Maintain self-improvement-loop?

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.

claude & codex plugins12 plugins