crisis-manager

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
  • Ingestion points: The skill reads untrusted log data from a file path provided via the --log argument in src/index.ts.
  • Boundary markers: There are no explicit boundary markers or 'ignore' instructions added to the log content before it is processed by the AI-driven analysis function.
  • Capability inventory: The skill has the capability to write files to the local system using safeWriteFile in src/index.ts based on the --out parameter.
  • Sanitization: The log content is read as a raw string and passed directly to generateRCAReport without sanitization or validation of the content's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 10:36 PM