tag-by-topic
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or behaviors were identified. The skill correctly implements its logic for tagging and ledger maintenance using local file operations.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from the 'seed body' and 'title'. However, the risk is mitigated by structured processing and strict regex validation of the output. 1. Ingestion points: Input 'seed body' and 'title' processed in SKILL.md. 2. Boundary markers: No explicit boundary markers are used for the input text. 3. Capability inventory: Reads from and appends to 'shared-context/topic-ledger.md'. 4. Sanitization: Employs regex validation (^[a-z0-9]+(-[a-z0-9]+)*$) for all proposed tag names before they are logged.
Audit Metadata