omh-live-incident-response
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process incident-related data provided by users or external signals, which constitutes an ingestion surface.
- Ingestion points: The workflow processes external information regarding blast radius, symptoms, and customer communication drafts within
SKILL.md. - Boundary markers: The skill establishes clear rules for evidence and status, such as distinguishing between 'prepared' and 'observed' states to prevent false claims of action.
- Capability inventory: The instructions involve formatting records for
live_incident_record/v1and coordinating with aconnector-operatortool for external notifications. - Sanitization: The skill explicitly instructs the agent to 'Never print customer records, credentials, tokens, or connection strings' in the incident timeline, mitigating data exposure risks.
- [SAFE]: The skill consists entirely of markdown documentation and workflow guidelines. It does not include any executable scripts, binary files, remote code execution patterns, or unverifiable external dependencies.
Audit Metadata