create-github-issues

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to create and publish issues to a remote repository. This action is the core functionality of the skill and is triggered only after the agent presents a proposed breakdown to the user for approval.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from existing GitHub issues and comments, which could potentially contain malicious instructions from external actors.
  • Ingestion points: The skill reads the full body and comments of issues referenced by the user in SKILL.md (Step 1).
  • Boundary markers: The instructions do not define explicit delimiters or warnings to the LLM to ignore instructions embedded within the fetched issue data.
  • Capability inventory: The agent has the ability to write to the repository's issue tracker via the gh CLI.
  • Sanitization: No automated sanitization of the input text is mentioned; however, the skill mandates a human-in-the-loop review step ("Quiz the user") in Step 4, where the user must approve the issue titles and descriptions before they are published.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 09:20 PM
Security Audit — agent-trust-hub — create-github-issues