The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted codebase content (source code and symbol names) to determine refactoring actions.
Ingestion points: Reads codebase structure and references through tools like gitnexus_query and gitnexus_context (SKILL.md).
Boundary markers: No delimiters or instructions to ignore instructions embedded within the code are present.
Capability inventory: The skill can execute file-system writes across multiple files via gitnexus_rename (SKILL.md).
Sanitization: No sanitization or validation of the ingested code content is documented before it influences the agent's plan.
[COMMAND_EXECUTION]: The skill instructs the user to execute the command npx gitnexus analyze to refresh the code index when it becomes stale.
[EXTERNAL_DOWNLOADS]: The skill relies on the npx package runner to download and execute the gitnexus utility, which is the core engine for the refactoring operations.

gitnexus-refactoring