omh-jev-review-gate
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits source code diffs to an external service via the
omh_jev_asktool for risk analysis. This behavior is documented and legitimate as part of the skill's primary function. It includes a mandatory notification to the user before sending data and a security filter that refuses diffs containing credential-like content to prevent accidental secret exposure. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of code diffs, creating a potential surface for indirect prompt injection.
- Ingestion points: Code diffs provided by the user for review, as handled in the
Proceduresection ofSKILL.md. - Boundary markers: The instructions do not define specific delimiters or isolation protocols for the diff content to prevent it from being interpreted as instructions.
- Capability inventory: The skill utilizes the
omh_jev_asktool, which performs network operations to analyze the provided diffs. - Sanitization: The skill performs a credential check on the diff content but does not sanitize the input for potential prompt injection patterns.
Audit Metadata