The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the execution of a local JavaScript file via Node.js to update the search index.
Evidence: The workflow section contains the instruction: node .gitnexus/run.cjs analyze.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as it ingests and processes untrusted project source code to perform automated refactoring.
Ingestion points: The impact(), query(), context(), and cypher() tools read code structure and content from the local workspace.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard natural language instructions that might be embedded in code comments or string literals within the processed files.
Capability inventory: The skill has high-impact capabilities, specifically the rename() tool which can perform multi-file edits across the repository, and the ability to trigger local script execution.
Sanitization: No sanitization or validation of the ingested code content is described before it is used to influence the agent's refactoring plan.

gitnexus-refactoring