sanity-check
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Indirect Prompt Injection] (SAFE): The skill defines a surface where local project data is ingested for verification.
- Ingestion points: Project files are read using the 'file: read' tool or searched via 'bash' tools (grep, find) to verify assumptions as described in the SKILL.md instructions.
- Boundary markers: No explicit delimiters are specified for the verification output.
- Capability inventory: Limited to file-system reads and searching (grep, find, test, read).
- Sanitization: No explicit sanitization of file content is described; the skill operates as a read-only logic layer within the agent's context.
- [Prompt Injection] (SAFE): No attempts to bypass safety filters or override system instructions were detected.
- [Data Exposure & Exfiltration] (SAFE): No hardcoded secrets, sensitive path access, or network communication patterns were identified.
Audit Metadata