migrate-pipeline
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of Harness pipeline configurations (v0 format) which represents an attack surface for indirect prompt injection.
- Ingestion points: Data enters the context via the
harness_getMCP tool output as described in Step 1 ofSKILL.md. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the fetched pipeline data as untrusted or to ignore embedded instructions within that data.
- Capability inventory: The skill has the capability to modify resources via the
harness_updateMCP tool, allowing the agent to write changes back to the Harness platform. - Sanitization: No sanitization, validation, or escaping of the fetched pipeline content is performed before the agent processes it for transformation.
Audit Metadata