The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the terminal tool to execute a suite of local Python scripts (pdf_read.py, pdf_split.py, pdf_page_image.py, extract_pymupdf.py, extract_marker.py) to handle complex PDF processing tasks.
[EXTERNAL_DOWNLOADS]: Instructions guide the agent to install missing Python dependencies such as pdfplumber, pypdf, pymupdf, pypdfium2, and marker-pdf if they are not already available. The skill explicitly recommends performing these installations within a task-local virtual environment to avoid polluting the main interpreter.
[DYNAMIC_EXECUTION]: The workflow involves dynamically resolving the paths of Python interpreters and creating runtime environments (python3 -m venv) to execute the required scripts.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process the full text of external documents, which are untrusted sources that may contain malicious instructions.
Ingestion points: Document paths or attachment references defined in SKILL.md.
Boundary markers: The skill uses a structured 'chunk ledger' and page-anchored ranges, but it does not provide explicit instructions to the model to ignore embedded commands or instructions within the extracted text.
Capability inventory: The agent has access to terminal (for script execution), read_file, vision_analyze, and the delegate_task tool for subagent management.
Sanitization: There is no mention of sanitizing or filtering the extracted document text before it is presented to the agent or its subagents.

omh-long-document-reading