github-conversation

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill contains a defensive security section that explicitly instructs the agent to identify and ignore potential prompt injection attempts (such as 'ignore previous instructions') embedded in third-party GitHub comments or descriptions.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing 'gh-llm', which is a specialized tool and GitHub CLI extension developed by the skill author (ShigureLab) to facilitate GitHub interactions.
  • [COMMAND_EXECUTION]: The skill utilizes the standard GitHub CLI ('gh') and the 'gh-llm' tool to perform its tasks. All command examples are restricted to the intended scope of reading and managing GitHub repository metadata and conversations.
  • [DATA_EXFILTRATION]: The skill interacts solely with the GitHub API for its documented purpose. It includes specific warnings to prevent the accidental posting of secrets or credentials found in third-party content.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 04:03 PM