task-generator
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (HIGH): Vulnerable to Indirect Prompt Injection through untrusted external data.
- Ingestion points: The skill ingests data from 'specs or requirements' provided by the user or external files.
- Boundary markers: There are no markers or instructions to isolate the specification text or tell the agent to ignore embedded instructions within the data.
- Capability inventory: The skill has the capability to write files to the local system (at
/tasks/tasks-[feature-name].md) and it generates shell commands likegit checkoutandnpx jestfor the user to run. - Sanitization: No sanitization or validation of the input spec is performed before it is translated into a task list. A malicious spec could include instructions that the agent converts into dangerous 'tasks' for a junior developer to execute.
Recommendations
- AI detected serious security threats
Audit Metadata