tdd
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to run test suites and verify acceptance criteria. Specifically,
SKILL.mdstates that 'every AC command passes from a shell' and directs the agent to 'run the existing suite' to confirm baselines. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and acts upon untrusted data from external sources such as task files, plans, issues, or feature requests.
- Ingestion points: Tasks, plans, and feature requests identified in
SKILL.md(Step 0: Frame the work). - Boundary markers: None identified; the skill does not specify delimiters or warnings to ignore instructions embedded in the external task data.
- Capability inventory: Shell execution of test commands and acceptance criteria.
- Sanitization: No input validation or sanitization of external commands is mentioned before they are passed to the shell.
Audit Metadata