The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local script scripts/opensearch_ops.py using the uv package manager to perform document ingestion, chunking, and quality evaluation.- [EXTERNAL_DOWNLOADS]: On its first execution, the skill downloads approximately 1.5 GB of AI models required for local document processing.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external documents (PDF, DOCX, PPTX, XLSX, etc.) which can serve as a vector for indirect prompt injection when the extracted text is processed by the agent.
Ingestion points: Reads unstructured data from user-supplied document files.
Boundary markers: The output is structured as JSONL, providing some separation between data fields and agent instructions.
Capability inventory: The skill has the ability to execute shell commands and write to the local file system.
Sanitization: The instructions specify validating the JSONL format before further use, though they do not describe specific text sanitization steps.

document-processing