signal-anchored-message
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of natural language instructions and test cases. No executable scripts, binaries, or package definitions are present.
- [PROMPT_INJECTION]: The skill processes untrusted data via inputs. Ingestion points: 'contact', 'company', and 'signal' in SKILL.md. Boundary markers: Absent. Capability inventory: No tool calls, file operations, or network requests are defined in SKILL.md or evals.json. Sanitization: Absent. The lack of capabilities prevents exploitation of the ingestion surface.
- [EXTERNAL_DOWNLOADS]: The documentation includes a reference to a GitHub repository for attribution purposes. This is a static link and does not involve automated software retrieval.
Audit Metadata