The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized network operations were detected in the skill instructions.
[DATA_EXPOSURE]: The skill requires access to local file paths to ingest PDF documents for processing. While this involves reading from the local file system, it is restricted to the paths provided to the ingest_documents and parse_pdf_structure tools, which is consistent with the skill's stated purpose of document analysis.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (PDF files) and transforms the content into Markdown and Base64 images for the agent's context.
Ingestion points: PDF files via ingest_documents and parse_pdf_structure tools.
Boundary markers: None explicitly defined in the extraction instructions to isolate extracted text from system instructions.
Capability inventory: The agent can read files and perform cross-document RAG queries via the consult_knowledge_graph tool.
Sanitization: No specific sanitization or filtering of extracted PDF content is mentioned. A malicious PDF could potentially contain instructions designed to influence the agent's behavior when the content is retrieved.

pdf-asset-extractor