SkillFeed

Turn failures and "I don't know what to use" moments into clear, ranked skill recommendations.

What it does

SkillFeed watches task context (goal + error/failure signal), builds targeted ClawHub queries, ranks candidate skills, and returns an actionable recovery plan.

• Detects common failure signals (API errors, timeout, retries, missing output)
• Generates layered search queries (broad → scenario → failure-specific)
• Recommends:
  • 1 primary skill
  • 2 alternatives
  • immediate next steps
  • fallback path
• Adapts response style for Claude Code / ChatGPT / Gemini

Typical use cases

• "Tweet publish failed, what should I use now?"
• "I need this automated but don't have time. Recommend a skill stack."
• "This workflow keeps timing out. Give me better options."

Output shape

• Goal
• Failure signal
• Primary recommendation
• Alternatives
• Immediate actions (3–5)
• Success checks
• Fallback
• Provider runbook (Claude Code / ChatGPT / Gemini)

Project structure

• SKILL.md – trigger rules + core workflow
• references/discovery-workflow.md – ranking and search strategy
• references/query-templates.md – query templates by scenario
• references/provider-adaptation.md – provider-specific output style

Install

clawhub install skill-feed

Notes

This project optimizes for practical recommendations (what to do next), not just listing possible skills.