pipeline-stage-definition-audit
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as an instructional and analytical framework for RevOps auditing. It does not utilize restricted tools, execute system commands, or perform network operations to untrusted domains. All external references target reputable organizations (e.g., Gartner, GitLab) or vendor-owned resources from the same author.
- [INDIRECT_PROMPT_INJECTION]: The skill includes a surface for ingesting user-provided business data, such as sales stage definitions and deal history, which is common for diagnostic tools. 1. Ingestion points: The Interview section and Artifact checklist (references/audit-intake.md) describe the ingestion of external process data. 2. Boundary markers: The instructions do not define specific delimiters for separating untrusted user data from the agent's core instructions. 3. Capability inventory: The skill possesses no dangerous capabilities (e.g., arbitrary command execution, network exfiltration, or file system writing) that could be exploited via the data ingestion surface. 4. Sanitization: No explicit input validation or sanitization logic is specified for the processed data.
Audit Metadata