factory-clean
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its processing of untrusted data to determine resource lifecycles.
- Ingestion points: The skill reads data from the 'run registry', 'local state file', and an external 'comment bus' (SKILL.md).
- Boundary markers: There are no explicit delimiters or instructions defined in the skill to treat the registry or comment bus data as untrusted data separate from instructions.
- Capability inventory: The skill has capabilities to scan file system paths, terminate processes and ports, and delete files/directories (SKILL.md).
- Sanitization: No specific sanitization, schema validation, or escaping of the ingested data is described before it is used to classify resources for deletion.
- [SAFE]: The skill implements strong security best practices for resource management. It defaults to a read-only state, explicitly protects resources with uncommitted changes or active heartbeats, and mandates a human-in-the-loop approval step where the exact commands are displayed before execution.
Audit Metadata