tdd
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read a
CONTEXT.mdfile (if it exists) to match the project's domain language and interface vocabulary. This creates a vulnerability surface where a project could include a maliciousCONTEXT.mdfile designed to influence the agent's behavior during the TDD process. - Ingestion points: The
SKILL.mdfile explicitly directs the agent to readCONTEXT.mdfor codebase exploration. - Boundary markers: The instructions lack guidance for the agent to use delimiters or to treat the contents of the context file as untrusted data, increasing the risk that the agent might follow embedded instructions.
- Capability inventory: The skill is intended to guide an agent in creating, modifying, and refactoring source code and tests, which involves high-privilege file system operations and code execution within the agent's environment.
- Sanitization: There are no requirements or mechanisms described to sanitize, validate, or filter the content ingested from the external context file.
Audit Metadata