The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local shell commands such as rg (ripgrep) for indexing repository content and npm/npx for running project-specific documentation build and validation scripts. These tools are standard for repository maintainers.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it is designed to ingest and process documentation and configuration files from the repository. 1. Ingestion points: Documentation source files in docs/, README.md, and configuration files like fern/docs.yml and docs/index.yml. 2. Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when reading the full content of documentation sections. 3. Capability inventory: Local command execution via rg and npm, as well as filesystem write access. 4. Sanitization: The process relies on manual review and automated route validation tests to verify the integrity of changes.
[SAFE]: The skill is authored by a trusted vendor and exclusively targets documentation within the vendor's repository. It does not perform unauthorized network operations, access sensitive credentials, or implement persistence mechanisms.

nemoclaw-maintainer-refactor-docs