tdd-workflows-tdd-refactor
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a workflow that is vulnerable to indirect prompt injection attacks because it handles external data without proper isolation.\n
- Ingestion points: The
$ARGUMENTSvariable inSKILL.mdacts as an entry point for untrusted code that is subsequently processed by thetdd-orchestratorsubagent.\n - Boundary markers: The prompt template lacks sufficient boundary markers or delimiters (such as triple backticks or XML tags) to effectively isolate the user-provided code from the agent's behavioral instructions.\n
- Capability inventory: The skill utilizes a powerful subagent,
tdd-orchestrator, which is designed to perform code modifications and execute tests, implying broad file system and shell command access.\n - Sanitization: No mechanisms for validating or sanitizing the input data are present, allowing potentially malicious instructions embedded in the input to be interpreted as agent commands.
Audit Metadata