digitaliza-data-extractor
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill processes untrusted external data such as scraped HTML and client-provided documents which could contain malicious instructions.
- Ingestion points: Specifically identifies
datos_extraier.md(scraped HTML/Markdown) and client folders as input sources inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided when processing these files.
- Capability inventory: The skill executes local directory listings and runs Python scripts for data processing.
- Sanitization: The provided
extract_colors.pyscript performs image processing but does not contain sanitization for text-based inputs mentioned in other steps. - Command Execution (SAFE): The skill instructs the agent to run local Python scripts (
extract_colors.py,extract_client_data.py) to automate data extraction. This is consistent with the skill's primary purpose and uses standard arguments. - Unverifiable Dependencies (SAFE): The color extraction script depends on the
Pillowlibrary, which is a standard and trusted package for image processing.
Audit Metadata