github-daily

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection.
  • Ingestion points: Fetches release notes and repository READMEs via 'gh release view' and 'gh api' in 'scripts/run_github_daily.py'.
  • Boundary markers: Absent. External content is summarized and placed directly into markdown reports without isolation markers.
  • Capability inventory: The skill executes shell commands via the GitHub CLI and writes files to the local filesystem.
  • Sanitization: Uses regex to strip formatting but does not filter for instructional keywords or bypass attempts.
  • [COMMAND_EXECUTION]: The skill executes shell commands to interact with GitHub.
  • Evidence: The script 'scripts/run_github_daily.py' uses 'subprocess.run' to invoke the GitHub CLI ('gh') for data retrieval and marketplace synchronization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 06:56 AM
Security Audit — agent-trust-hub — github-daily