The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill suggests installing several well-known and reputable Python packages from standard registries, including pdfplumber, pypdf, pillow, pytesseract, pandas, pdf2image, jsonschema, and openpyxl. It also references official system package installation for Tesseract OCR via brew (macOS) and apt-get (Ubuntu), as well as the official Tesseract Windows binaries repository on GitHub.
[COMMAND_EXECUTION]: The skill documentation and examples demonstrate the use of subprocess.run and CLI commands to execute its own internal scripts (e.g., scripts/analyze_form.py, scripts/fill_form.py). This is standard functionality for a script-based processing toolkit and does not involve arbitrary command execution or suspicious shell piping.
[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8).
Ingestion points: The skill ingests untrusted data from external PDF files via extraction scripts in SKILL.md, FORMS.md, TABLES.md, and OCR.md.
Boundary markers: No explicit instructions or delimiters are provided to the agent to distinguish between the extracted PDF data and control instructions.
Capability inventory: The skill possesses the capability to execute subprocess calls (subprocess.run) and perform file system operations (read/write).
Sanitization: There is no evidence of sanitization or filtering of the extracted PDF content before it is processed by the agent. This represents a potential attack surface if a malicious PDF contains instructions designed to manipulate the agent's behavior after text extraction.

PDF Processing Pro