proof-theory
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts mathematical formulas as arguments to shell-executed scripts, creating a potential surface for indirect prompt injection. This is a characteristic of logic-solving tools and is considered low risk in this context.
- Ingestion points: Mathematical expressions passed to
z3_solve.pyandmath_scratchpad.pyinSKILL.md. - Boundary markers: None.
- Capability inventory:
Bashtool used to execute scripts via a runtime harness. - Sanitization: Not explicitly defined.- [SAFE]: No malicious patterns such as credential theft, obfuscation, or remote code execution were found. The skill operates on local scripts using a controlled execution environment.
Audit Metadata