monorepo-navigator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the implementation of workspace-aware configuration using CLAUDE.md files. These files provide instructions that guide agent behavior based on the specific package context within a monorepo.\n
- Ingestion points: CLAUDE.md and package.json files located within the project workspaces.\n
- Boundary markers: The skill recommends using explicit task scoping rules but does not include generic markers to delimit untrusted content.\n
- Capability inventory: Shell command execution (Turbo, Nx, pnpm) and file system read/write operations.\n
- Sanitization: The monorepo_analyzer.py script uses standard JSON parsing to handle package metadata.\n- [COMMAND_EXECUTION]: Documentation provides a utility script snippet using child_process.execSync to run pnpm commands for dependency visualization.\n- [DYNAMIC_EXECUTION]: The reference material includes an example of generating a local JavaScript script (gen-dep-graph.js) from a template and executing it to build a dependency graph.
Audit Metadata