workflow-log-bug

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified. The skill incorporates strong restrictive guidelines to limit the agent's actions, explicitly forbidding codebase exploration and network access.
  • [PROMPT_INJECTION]: The skill ingests user input regarding bug symptoms and writes it to the local file system, which constitutes an indirect prompt injection surface. 1. Ingestion points: User conversation context and symptoms provided by the user. 2. Boundary markers: Absent. 3. Capability inventory: Bash (restricted to mkdir -p) and file writing to a specific subdirectory. 4. Sanitization: No explicit validation or sanitization is mentioned for the user-supplied symptoms or generated slugs.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 06:53 AM