understand-issue

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources.
  • Ingestion points: The skill retrieves issue titles, bodies, and comments using gh issue view and gh issue list.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat fetched issue content as untrusted data or to ignore instructions embedded within that content.
  • Capability inventory: The skill can write to the filesystem, execute git and gh commands, and post comments to GitHub issues.
  • Sanitization: No sanitization or validation of the fetched GitHub data is performed before it is used to generate summaries or contribute to the diagnosis.
  • [COMMAND_EXECUTION]: The skill uses standard CLI tools (git, gh) to perform repository analysis and GitHub interaction. This execution is scoped to the skill's primary purpose of issue investigation and does not include arbitrary or high-privilege command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 12:02 AM
Security Audit — agent-trust-hub — understand-issue