setup
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands, specifically
cp, to copy scaffolding templates and scripts from the skill package into the target repository's.agents/directory.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific metadata and user-defined instructions from files likePROJECT.mdandINSTRUCTIONS.mdto customize the setup process.\n - Ingestion points: Reads configuration data from
.agents/docs/PROJECT.md,.gitignore, and the rootAGENTS.mdfile.\n - Boundary markers: Uses markdown headers and table structures as delimiters to parse structured information from documents.\n
- Capability inventory: Includes file system read/write operations and shell command execution for directory management.\n
- Sanitization: The included utility scripts (
cooking.mjs,spec-files.mjs) implement path validation and segment checking to mitigate directory traversal risks when processing file paths.\n- [DYNAMIC_EXECUTION]: The skill copies a set of Node.js utility scripts into the workspace and subsequently executes them using thenoderuntime to verify the setup state and manage internal workflows.
Audit Metadata