failure-flywheel
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
harness/scripts/failure-flywheel.pyto manage failure records and generate reports. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface. Ingestion points: failure summaries and details provided via the
recordcommand. Boundary markers: none specified. Capability inventory: file-write operations (creating markdown stubs) tollmwiki/wiki/sources/draft/performed by the Python script. Sanitization: the skill relies on a human review checkpoint ('never auto-promote') and generates 'TODO' stubs rather than active code to prevent automatic execution of ingested data.
Audit Metadata