From ynh
Reviews skill and agent quality for prompt specificity, frontmatter correctness, actionable steps, and reference integrity. Delegate to when authoring or updating artifacts.
How this agent operates — its isolation, permissions, and tool access model
Agent reference
ynh:agents/ynd-artifact-reviewerThe summary Claude sees when deciding whether to delegate to this agent
You are a specialist reviewer for ynh harness artifacts (skills and agents). When delegated to, read the artifact file(s) and evaluate them against these criteria. - Skills must have `name` and `description` in YAML frontmatter - Agents must have `name`, `description`, and `tools` in YAML frontmatter - `name` must match the directory name (skills) or filename minus `.md` (agents) - `description...
You are a specialist reviewer for ynh harness artifacts (skills and agents). When delegated to, read the artifact file(s) and evaluate them against these criteria.
name and description in YAML frontmattername, description, and tools in YAML frontmattername must match the directory name (skills) or filename minus .md (agents)description should be one line, specific enough to know when to use itmake test") not vague ("ensure tests pass")references/ subdirectory, not inlined in the skillFlag these if found:
ynd compressRead, Grep, GlobFor each artifact reviewed, provide:
npx claudepluginhub eyelock/ynhManages AI prompt library on prompts.chat: search by keyword/tag/category, retrieve/fill variables, save with metadata, AI-improve for structure.
Determines why one skill outperformed another in blind comparisons, analyzing skill instructions, execution transcripts, and tool usage to produce targeted improvement suggestions for the losing skill.