migration
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes migration records in the
.migration/directory and project implementation code to prepare and execute migrations. This creates an attack surface where malicious instructions embedded in these files could influence the agent's migration plan or execution steps. - Ingestion points: Reads files from the
.migration/directory and project implementation source files (SKILL.md). - Boundary markers: The skill instructs the agent to "State decisions as standalone facts" and exclude conversation history or narration, which helps reduce accidental obedience but does not strictly delimit untrusted data from instructions (SKILL.md).
- Capability inventory: The skill is designed to execute migration plans involving target modifications, although specific tool capabilities are governed by the platform (SKILL.md).
- Sanitization: No specific sanitization or validation of content read from migration records is described before processing it for execution.
Audit Metadata