tdd
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially contain malicious instructions, creating a surface for indirect prompt injection.
- Ingestion points: Test execution output (stdout/stderr) from local shell commands and strategic suggestions from the
oh-my-claudecode:test-engineersub-agent (likely a resource provided by the author). - Boundary markers: The skill defines a structured output format for the TDD cycle phases but does not specify delimiters or "ignore instructions" wrappers for the ingested external content.
- Capability inventory: The skill utilizes shell command execution (
bash) to run the project's test suite. - Sanitization: No explicit sanitization or filtering is applied to the ingested data, though the instructions mandate that the agent must "critically evaluate" and "never blindly adopt" external suggestions.
- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands to run tests, which is a necessary capability for its intended purpose but represents a standard security surface for the agent.
Audit Metadata