Skill Ops Suite

Runtime-neutral skill operations suite. Input a skill directory, slug, marketplace goal, or release plan; output quality checks, safety boundaries, naming/summary improvements, install-conversion diagnosis, duplicate/merge/redirect recommendations, and a release checklist. Review-only unless the user explicitly approves publishing or destructive changes.

Use this flagship skill when the user is creating, auditing, publishing, growing, or maintaining agent skills across different agent runtimes and skill marketplaces.

User Promise

Input a skill directory, skill slug, market goal, or release plan. Output quality and installability checks, safer name/display name/summary/keywords, security boundary review, duplicate/merge/redirect recommendations, and release checklist.

Safety Boundaries

• Do not publish, delete, rename, archive, or run destructive commands without explicit user approval.
• Do not upload secrets, credentials, private customer data, or sensitive raw logs.
• High-risk domains such as legal, medical, payments, identity, or browser automation require clear disclaimers and stop rules.

Example Prompts

1. Review this skill and tell me why users might not install it.
2. Rewrite this skill summary using input/output/safety-boundary format.
3. Find duplicate skills in this set and recommend merge or redirect actions.
4. Prepare a release checklist for publishing this agent skill.
5. Audit this shopping skill for login, checkout, payment, and privacy boundaries.