subagent-driven-development
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a logical multi-agent orchestration pattern for executing independent development tasks. No malicious behavior was identified.
- [COMMAND_EXECUTION]: Subagents are instructed to perform typical developer tasks, such as implementing features, writing tests, and committing code. These operations are performed using the agent's built-in tools within the local development environment.
- [PROMPT_INJECTION]: There are no patterns found that attempt to bypass AI safety filters or override system-level instructions. The instructions provided to subagents are focused on task execution and critical review.
- [DATA_EXFILTRATION]: No network calls or attempts to access sensitive files (like credentials or configuration folders) were detected. The skill interacts exclusively with project-specific code and implementation plans.
- [INDIRECT_PROMPT_INJECTION]: The skill processes implementation plans and task descriptions which are interpolated into prompts for subagents. While this presents a theoretical surface for indirect injection (Cat 8), the risk is low as it is the primary functional purpose of the orchestrator, and the subagents are directed to perform reviews that mitigate potential non-compliance or malicious content within those tasks.
Audit Metadata