An evaluation and evolution tool for Agent Skills.
skill-up is a tool for evaluating and improving Agent Skills through declarative test cases, multiple agent engines, configurable judges, and structured reports. Developers and agent builders use it locally or in continuous integration to measure skill quality and iterate on failures. The catalogue entries are skills and instructions that can be evaluated or evolved with it.
This repository also configures its own agents. See what skill-up tells them →
Hook Claude Code ✓ vendor
One of 4 in hooks.json
Runs when you submit a prompt, before the agent sees it, executing observer.py via python3. From alibaba/skill-up.
Hook Claude Code ✓ vendor
One of 4 in hooks.json
Runs after a tool call finishes for ^mcp__skill_up_observer__(mark_skill_invocation, attach_skill_evidence and record_skill_feedback)$ tool calls, executing observer.py via python3. From alibaba/skill-up.
Hook Claude Code ✓ vendor
One of 4 in hooks.json
Runs when the agent finishes a response, executing observer.py via python3. From alibaba/skill-up.
Hook Claude Code ✓ vendor
One of 4 in hooks.json
Runs on the Interrupt event, executing observer.py via python3. From alibaba/skill-up.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: