tml-capture
Pass
Audited by Gen Agent Trust Hub on Feb 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The analysis did not reveal any malicious patterns, obfuscation, or unauthorized data access. The skill manages data transformation through a series of specialized sub-skills with explicit human verification gates defined in the pipeline configuration.
- [PROMPT_INJECTION]: The skill processes untrusted external data, creating a potential indirect injection surface. 1. Ingestion points: 'context-dump' (pasted text, file contents, wikis). 2. Boundary markers: 'inferences-document' distinguishes between 'Stated', 'Inferred', and 'Gap' content. 3. Capability inventory: 'the-agent-brief' outlines specific read/write connectors and role boundaries. 4. Sanitization: All critical stages require a 'human-gate' for validation. Given that processing external context is the primary skill purpose and it is gated by human review, this surface is considered SAFE.
Audit Metadata