LLM-driven session retrospection skill: detects friction in agent sessions and materializes learnings into correct destinations (user-memory, project-rules, skill PRs, checkpoints, harness artefacts)
Python Multi-Objective Simulation Optimization: a package for using, implementing, and testing simulation optimization algorithms.
An end-of-session /retro skill for Claude Code: the agent reviews its own session, keeps only lessons that clear an explicit bar, promotes repeat lessons into hooks, prunes to a budget, and gates anything that changes its behavior on a human.
mentor — a session-insights skill for AI coding agents. This skill reads your local Claude Code and OpenAI Codex history and writes an /insights-style HTML report on how you work: what you build, where you lose time, and concrete fixes. An agent skill for Claude Code, Codex, and any skills-capable agent.
Aplicacion desarrollada con python que permite gestionar (hacer backup y restore) a nuestras bases de datos.
Prompt style library with honesty guardrails.
Find what slowed the PR down, and fix it for the next one. Reads the whole Claude Code and Codex conversation after a merge.
Task Evidence Lab - 本地任务证据与质量复盘系统
Entropy, compression, and information-theoretic rune tools.
Agent Skill for designing visually engaging agile/scrum retrospective boards (Draw.io, Mermaid, PNG/SVG, Microsoft Whiteboard) and turning completed retros into summaries, action items, and Jira/Confluence-ready output.
Feynman-style numerical computing lessons with runnable examples.
Strategic thinking as installable infrastructure — 21 skills, 13 seed patterns, 5-phase pipeline. Give your AI agent the discipline to explore, decide, execute, and learn.
本地经验复盘与复用。核心特色：跨设备/跨客户端的经验整合 + 整档案的备份与恢复；经验只存本机、不上传，明文可编辑、可审计、可按需合并。Local-first experience layer for AI agents: merge your experience across devices and clients, back it up, never upload. 关键词：经验复盘、经验沉淀、长期记忆、个性化、跨 Agent 迁移。
Agent-agnostic skill for recalling verified lessons and handling unproductive retries with lightweight reminders.
A /retro skill for Claude Code: milestone retrospectives that carry scoped, size-capped lessons across projects
From-scratch neural network and battle outcome trainer.
Local-first multi-agent work evidence aggregation, structured summaries, and standalone HTML presentations
An honest, step-by-step post-mortem of six autonomous Polymarket trading bots built and killed over three months. Zero edge found — and why that 'no' was the only real win. Code, autopsy, and lessons. Built with Claude Fable 5.
Evolutionary strategy and particle swarm optimizer demos.
Sprint retro for your Claude Code sessions. See what worked, what didn't, and what to change.
To associate your repository with the retrospective topic, visit your repo's landing page and select "manage topics."