switching-projects
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read package.json and README.md from the target project directory to orient itself after switching. This represents an ingestion point for untrusted data that could influence subsequent agent behavior.
- Ingestion points: Project metadata files like package.json and README.md (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: Shell commands (ls, git status) and workspace management tools (move_agent_to_root).
- Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill utilizes shell commands such as ls to list directories in common development paths and git status to verify project state. These commands operate on the local file system.
Audit Metadata