tsg-flow-product
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a Python validation script (
validate_state.py) located in a related repository path (tsg-flow-next). This command is used to verify the consistency of the documentation state and is a standard part of the vendor's local development workflow. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an ingestion surface for indirect prompt injection by reading project source code, Architecture Decision Records (ADRs), and existing documentation to extract business logic and decisions.
- Ingestion points: Local project source code, files in
docs/adr/, and existing documents indocs/product/. - Boundary markers: Absent; there are no explicit instructions to the agent to delimit or ignore instructions that might be embedded within the ingested code or documents.
- Capability inventory: The agent possesses file-writing capabilities (creating markdown and JSON artifacts in
docs/product/) and the ability to execute the specific Python validation script mentioned above. - Sanitization: No sanitization or content validation is performed on the ingested data before it is used to populate documentation templates.
Audit Metadata