The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/repomix_batch.py uses subprocess.run to execute the repomix CLI and npx for repository processing. This is the intended functionality of the batch utility.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download of the repomix package from the NPM registry via npx when remote repositories are targeted. It also references the official yamadashy/repomix GitHub repository.
[PROMPT_INJECTION]: The skill is designed to aggregate codebase content for LLM consumption, which creates a surface for indirect prompt injection if the source code contains malicious instructions.
Ingestion points: File contents from local or remote repositories processed by scripts/repomix_batch.py.
Boundary markers: Structural delimiters such as XML tags, Markdown headers, and JSON keys are used to separate individual file contents.
Capability inventory: The agent consuming the packaged output may have capabilities like file system access, network operations, or tool execution depending on its specific configuration.
Sanitization: The tool integrates Secretlint to detect sensitive data (API keys, secrets), though it does not specifically sanitize natural language instructions within the code.

repomix