testing
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches skill configuration and instructions from the official braintrustdata GitHub repository, which is a recognized source for AI development tools.\n- [COMMAND_EXECUTION]: Executes transformation coverage reports and SDK validation tests using cargo run and pnpm commands within the local project environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes transformation test cases, creating a potential surface for indirect instructions.\n
- Ingestion points: Processes JSON test cases located in payloads/cases/ and transform_errors.json.\n
- Boundary markers: None identified; the tool relies on structured JSON schemas for validation.\n
- Capability inventory: Includes shell command execution via cargo and pnpm tools.\n
- Sanitization: Employs schema validation at every transformation step to ensure data integrity and prevent malformed input processing.
Audit Metadata