create-pipeline
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes codebase files (e.g., package.json, go.mod, pom.xml) to auto-detect build and deployment settings, as detailed in references/codebase-analysis.md. This ingestion of untrusted data represents an attack surface where malicious instructions hidden in these files could influence the generated pipeline.
- Ingestion points: Scanning of local project files such as package.json, go.mod, pom.xml, and requirements.txt during the analysis phase.
- Boundary markers: The instructions require the agent to "Confirm detected settings with the user" and "Ask about anything that couldn't be auto-detected," providing a human-in-the-loop validation step.
- Capability inventory: The skill can invoke harness_create and harness_update MCP tools to modify pipeline configurations.
- Sanitization: The instructions do not specify any mechanisms for sanitizing file content or explicitly ignoring instructions embedded within the processed codebase.
Audit Metadata