task-breakdown
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill utilizes the
$ARGUMENTSplaceholder to capture user input, which is then used to generate task and subtask JSON files. These files provide the operational context and success criteria for other agents (such asCoderAgentorTestEngineer). If the input contains malicious instructions, they may be propagated into the workflows of these subagents. - Ingestion points: The
$ARGUMENTSparameter in theSKILL.mdfile. - Boundary markers: Absent. There are no delimiters or warnings to prevent the agent from following instructions embedded within the user-provided text.
- Capability inventory: The skill writes JSON files to the
.tmp/tasks/directory. These files govern the actions of other agents with capabilities for file system modification and code execution. - Sanitization: Absent. The skill does not include steps to sanitize or validate the user-provided input before it is used to generate task descriptions and criteria.
Audit Metadata