gh-activity-summary

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Indirect Prompt Injection (LOW): This skill has an attack surface for indirect prompt injection. It aggregates data from external sources (GitHub commit messages, PR titles, and issue descriptions) that could contain malicious instructions designed to influence an AI agent processing the final report.
  • Ingestion points: GitHub API via gh search in activity-report.sh.
  • Boundary markers: The script uses markdown headers but does not include explicit instructions to the AI to ignore content within the data sections.
  • Capability inventory: No dangerous local capabilities (like exec or eval) are present in the script; it only outputs text.
  • Sanitization: Commit messages and titles are extracted and printed without sanitization of control characters or instruction-like text.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:49 PM