export-open-issues

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the official GitHub CLI (gh) to fetch repository data and utilizes standard system commands (date, mkdir) for timestamping and managing local audit directories.
  • [EXTERNAL_DOWNLOADS]: Data is retrieved from well-known and trusted platforms (GitHub, Linear, and Jira) through their respective official tools and APIs.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted issue data (titles, labels, and metadata) which represents a potential injection surface, though the logic is confined to report generation.
  • Ingestion points: Issue and PR data retrieved from external trackers in SKILL.md.
  • Boundary markers: Data is structured within markdown tables, but no specific 'ignore instructions' delimiters are applied.
  • Capability inventory: Local file system creation (mkdir) and CLI execution (gh) defined in SKILL.md.
  • Sanitization: No explicit sanitization or filtering of external issue content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 11:30 AM