test-driven-development
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as
npm testto verify test failures and successes. This is a functional requirement for the TDD lifecycle described in the instructions. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source code and test files during the TDD cycle, which creates a surface for potential indirect prompt injection if those files contain malicious instructions meant to influence the agent's behavior.
- Ingestion points: The agent is instructed to read, write, and verify code and test files throughout the 'Red-Green-Refactor' process.
- Boundary markers: The skill provides strong instructional constraints ('The Iron Law', 'The Final Rule') but does not specify delimiters or security boundaries to distinguish between its own instructions and the content of the files it processes.
- Capability inventory: The skill leverages
write_filefor code creation and shell execution for running tests via NPM. - Sanitization: There is no evidence of content sanitization or validation before the agent processes or executes the code provided in the project files.
Audit Metadata