Benchmarks for Compose by Focus: Scene Graph-based Atomic Skills
APPL is an agentic robot learning framework connecting policy learning with agent-based skill selection and composition through structural priors for generalization from a few demonstrations to new scenes and task compositions.
Platform-aware skill composition layer for AI coding tools
Voice-controlled MCP hardware gateway for UNIHIKER K10: reusable skills, composition, and Codex-generated programs.
SkillWeave: A Semantic Composition Framework for Multi-Agent Skill Orchestration in Enterprise Environments
Capability-Augmented Generation through Reusable Graph — a research framework for accumulating and composing executable LLM skills.
A lightweight, interactive meta-skill for composing AI agent skills into task-specific workflows.
Merge, compile and update AI agent skills with provenance: every source read in full and scanned, licences checked, each conflict named in MERGE.md, a lockfile per commit, and an eval gate that refuses a merge that loses. Claude Code, Codex, Copilot.
Code & data for arXiv:2604.26689 — atomic-probe governance for skill updates in compositional robot policies
学→超→破→用：把开源 agent skill 集合真正用起来的方法论（求合作防蒸馏脱敏版 · L3 配方层合作交付）
To associate your repository with the skill-composition topic, visit your repo's landing page and select "manage topics."