From thought to skill. From signal to structure.
-
Updated
Sep 2, 2026 - Python
From thought to skill. From signal to structure.
[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Record desktop workflows once and compile them into reusable AI-agent skills for reliable replay.
University for AI agents. 92 courses, 4400+ scenarios, any model via OpenRouter. Auto-training loops generate per-model SKILL.md documents. Works with Claude Code, OpenClaw, Cursor, Windsurf. No fine-tuning required.
AI assistant that learns your workflows — auto-generates skills, routes to the cheapest LLM, and replays every execution. Telegram · Slack · Discord · Matrix. One Rust binary, zero setup.
ARCHIVED — Skill generation tool. Absorbed into Gravity-Omega tool ecosystem.
Turn development lessons into reusable skills — automatically. Analyze patterns, propose skills, sync improvements, check health.
The skill that writes skills — a secure compiler pipeline that auto-generates OpenClaw skills from observed user patterns
Convert OpenCode shared sessions into reusable, publishable AI agent skills with TRACE validation, security checks, and ZIP packaging.
Create custom domain expert skills through interactive dialogue — hire, summon, and manage your expert team for Claude Code
Progressive-disclosure reflector for OpenCode — automatically analyzes conversations to discover reusable patterns and generate skills/rules
Build self-updating skill trees for AI coding agents from codebases and PRDs to keep agents in sync with your code
The world's first skill-forging engine built for AI Agents in film & video production. 全球首个面向 AI Agent 的影视创作技能生成引擎。
To associate your repository with the skill-generation topic, visit your repo's landing page and select "manage topics."