analyze-project

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes session logs and metadata from ~/.gemini/antigravity/brain/, providing an entry point for adversarial instructions to influence the agent's analysis. \n- Ingestion points: Conversation logs and artifacts are read during Step 1 and Step 2. \n- Boundary markers: The instructions lack explicit delimiters or warnings to treat the session data as untrusted. \n- Capability inventory: The skill reads local files and writes diagnostic reports to the project directory. \n- Sanitization: Content from analyzed sessions is interpolated into reports without evidence of sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 08:48 AM
Security Audit — agent-trust-hub — analyze-project