db-schema-designer
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and analyze external artifacts like product requirements, protobuf files, and existing schema definitions, creating a potential surface for indirect prompt injection if these inputs contain malicious instructions.
- Ingestion points: The skill reads external documentation, proto files, and existing database schemas during the 'Explore inputs' and 'Phase 1' steps described in SKILL.md.
- Boundary markers: There are no specific delimiters or instructions provided to the agent to distinguish between data and potentially embedded instructions within the ingested files.
- Capability inventory: The skill uses tools to create tasks (TodoWrite) and writes design briefs to the local file system (e.g., prd/DDL.md).
- Sanitization: No explicit sanitization, validation, or escaping rules are defined for handling the content extracted from external documents.
- [SAFE]: No malicious code, obfuscation, network exfiltration, or persistence mechanisms were detected. The design process is entirely incremental and requires explicit human approval before any documentation is finalized or handed off to downstream skills.
Audit Metadata