skills/spotdraft/oncall-investigator-skills/approval-watermark-template-execution-mismatch/Gen Agent Trust Hub
approval-watermark-template-execution-mismatch
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structured debugging steps using internal vendor infrastructure, including the SpotDraft Admin API and BigQuery datasets. All endpoints identified are part of the vendor's authorized ecosystem.
- [SAFE]: References to external services like Slack and Jira are used for searching incident history and tracking task status (e.g., Jira SPD-42223), which are expected developer and support workflows.
- [SAFE]: While the skill involves processing data from external support tools (Category 8: Indirect Prompt Injection surface), it is restricted to internal diagnostic functions. Evidence Chain: (1) Ingestion points: Slack threads and Jira tickets; (2) Boundary markers: None; (3) Capability inventory: BigQuery SQL execution and Admin API access; (4) Sanitization: None. This surface is necessary for the skill's primary purpose and does not facilitate malicious activity.
Audit Metadata