Subagent-Driven Development
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a documentation-based guide that defines a process for the agent to coordinate subtasks. It does not ship with any external dependencies, scripts, or binary executables.
- [INDIRECT_PROMPT_INJECTION]: The workflow involves reading a plan file as an input for task generation. While this represents a data ingestion surface, the instructions are designed for a standard development lifecycle (implementation, testing, and review), posing no significant security risk beyond the inherent behavior of processing project documentation.
- [COMMAND_EXECUTION]: The skill describes a process where subagents implement code and run tests. This is a standard functional capability for an engineering-focused AI agent and is executed within the context of defined development tasks without evidence of malicious command injection.
- [REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts from untrusted sources. All operations described are local to the development environment and based on internal agent tools.
Audit Metadata