omh-relational-db
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely advisory and acts as a planner. It does not possess capabilities to execute shell commands, perform network requests, or access the file system beyond reading its own reference files.
- [SAFE]: Explicit safety instructions are provided to prevent the leakage of sensitive data. It warns users and the agent to never include connection strings, credentials, or customer rows in any plans or handoffs.
- [SAFE]: The technical guidance provided in the reference file (
references/engine-lock-tables.md) promotes security best practices, such as using non-blocking DDL operations and warning against the risks of runningEXPLAIN ANALYZEon production primary databases for write queries. - [SAFE]: The use of vendor-specific commands (e.g.,
omh runtime record) is documented for internal tracking and does not involve unauthorized execution of remote code or privilege escalation.
Audit Metadata