agent-skill-rules
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown documentation files that define a structural standard for agent skills. It does not perform any sensitive operations or access restricted resources.
- [NO_CODE]: The skill contains no executable scripts (e.g., Python, JavaScript, or Bash). All logic is declarative and instructional, intended to guide the behavior of an AI agent when managing skill files rather than executing code directly.
- [EXTERNAL_DOWNLOADS]: The documentation references external URLs such as 'agentskills.io', 'skills.sh', and a GitHub repository for the 'skills-ref' validation tool. These are provided as informational links for the user or agent to reference for specification and manual verification; the skill itself does not automate the downloading or execution of these resources.
- [PROMPT_INJECTION]: While the skill contains detailed instructions on how an agent should 'audit', 'refactor', or 'migrate' other skills, these are meta-rules for governance. There are no attempts to bypass safety guardrails, hide instructions, or override the agent's core system prompts.
Audit Metadata