tdd-workflows-tdd-refactor
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interpolates user-provided content via the
$ARGUMENTSvariable directly into a prompt for a sub-agent (tdd-orchestrator) without using boundary markers or sanitization. This allows malicious instructions embedded in the code-to-be-refactored to potentially influence the behavior of the sub-agent. - Ingestion points: Found in
SKILL.mdwhere$ARGUMENTSis interpolated into the usage prompt. - Boundary markers: Absent; the user input is appended directly to the end of the instruction block without delimiters like triple backticks or XML tags.
- Capability inventory: The skill utilizes the
Tasktool for sub-agent orchestration (tdd-orchestrator). - Sanitization: No input validation or explicit instructions to ignore embedded directives are present.
Audit Metadata