The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes various office and image file formats which may contain embedded instructions, creating a surface for indirect prompt injection.\n- Ingestion points: The skill ingests Word, Excel, PPT, HTML, TXT, CSV, RTF, and image files as defined in SKILL.md.\n- Boundary markers: There are no instructions for the agent to use specific delimiters or to disregard potential instructions within the source files.\n- Capability inventory: The agent is authorized to prepare network requests and resolve local credentials.\n- Sanitization: No sanitization or validation of file content is specified before the conversion process.\n- [COMMAND_EXECUTION]: The skill generates instructions for interacting with ComPDF API endpoints. It maps user requests to specific operations like 'word-to-pdf' and 'html-to-pdf' using referenced documentation.\n- [DATA_EXFILTRATION]: The skill manages authentication via a local 'api_key' file. It provides specific instructions to ensure the key is handled as private runtime state, prohibiting its inclusion in logs, examples, or final outputs, and restricting the agent to reading only the first line of the file.

compdf-documents-to-pdf