The Agent Skills Directory

[SAFE]: A comprehensive analysis of the skill's documentation and the script scripts/dpa_clause_check.py reveals no malicious intent. The skill performs deterministic legal analysis without making network requests, accessing sensitive credentials, or using obfuscated code. It adheres to the author's stated purpose of providing an offline, local-only compliance tool.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to ingest and analyze untrusted external data in the form of legal contracts.
Ingestion points: The skill processes user-supplied DPA contract text and RCP records through its analysis workflow.
Boundary markers: No specific delimiters or instructions to ignore embedded commands (e.g., within contract comments) are defined in the instructions.
Capability inventory: The skill is authorized to use Bash for running the local Python validator and Read for file operations.
Sanitization: No evidence of input sanitization or filtering of the processed text was found.
Note: This is a characteristic of document-processing tools and, given the deterministic nature of the code, does not elevate the risk level for this specific implementation.

rodo-ropa-dpa-pl