The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill extracts text content from external .docx files and presents it to the agent, creating a surface for indirect prompt injection attacks where a document's content could hijack agent behavior.\n- Ingestion Points: scripts/inspect_docx.py and scripts/edit_docx.py read text content from user-provided documents.\n- Boundary Markers: The instructions do not define clear boundaries or warn the agent to ignore potential instructions embedded within the document text.\n- Capability Inventory: The skill possesses the ability to write to the file system (doc.save()) and execute shell commands (unzip, zip).\n- Sanitization: No sanitization or filtering is performed on the text extracted from the documents.\n- [COMMAND_EXECUTION]: The SKILL.md instructions recommend that the agent use shell commands (unzip and zip) to manually patch document XML files when high-level libraries are insufficient. While these are appropriate tools for the task, they grant the agent direct command-line capabilities for file manipulation.

docx