The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill extracts text from user-provided PDF files, creating a surface for indirect prompt injection attacks where malicious instructions could be embedded in document content.
Ingestion points: The script scripts/pdf-extract.js reads raw file data using fs.readFileSync(args.file).
Boundary markers: No specific boundary markers or instructions to ignore embedded content are used when outputting extracted text to the agent.
Capability inventory: The script is limited to local file reading and standard output. It contains no network operations, file-writing capabilities, or command execution logic.
Sanitization: The script performs basic character filtering to remove non-printable characters but does not sanitize the text for potentially malicious instructions.

pdf-extractor