prune
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill executes external analysis tools (e.g., knip, ruff, vulture) using ephemeral runners such as npx, uvx, bunx, and go run. Additionally, it discovers and executes verification commands like tests and build scripts defined in the project's manifests (package.json, Makefile, CI configs). \n- [DYNAMIC_EXECUTION]: The skill constructs and executes shell commands at runtime based on the project environment and configuration files to perform its refactoring and verification tasks. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes project-local files like docs/prune/keep.md and manifests to determine its scope and constraints, creating a vulnerability surface. \n
- Ingestion points: docs/prune/keep.md, package.json, Makefile, and CI configuration files. \n
- Boundary markers: The skill does not use specific delimiters or instructions to ignore potential commands embedded in these configuration files. \n
- Capability inventory: The skill can execute shell commands, write to the filesystem, and perform Git operations. \n
- Sanitization: There is no evidence of validation or sanitization for the commands or configuration values read from the project environment.
Audit Metadata