omh-live-incident-response
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data regarding incident symptoms and blast radius. Ingestion points: User-provided descriptions of outages and system behaviors in SKILL.md. Boundary markers: Implements a structured record format (live_incident_record/v1) to encapsulate data. Capability inventory: No dangerous capabilities such as subprocess execution, network operations, or arbitrary file access were found. Sanitization: Safety rules explicitly prohibit the inclusion of credentials, tokens, or customer records within the incident timeline.
- [SAFE]: The skill contains specific safety directives that prevent the exfiltration or exposure of sensitive information like tokens or credentials during the incident command process.
- [NO_CODE]: The skill consists entirely of instructional markdown files and does not include any executable scripts or binary files.
Audit Metadata