The Agent Skills Directory

[SAFE]: The skill performs text analysis on legal documents as described. All external references (GitHub, MCP connectors) are well-documented and belong to the author or trusted open-source legal tech projects. No unauthorized network operations or system modifications are present.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (court rulings), which creates an indirect prompt injection surface. This is a standard risk for document analysis tools.
Ingestion points: The skill ingests text from Polish court databases (SAOS, EUR-Lex) or user-provided text files using the Read tool.
Boundary markers: The instructions do not define explicit delimiters to isolate the legal text from the agent's instructions during processing, though it uses quotes for the final output formatting.
Capability inventory: The skill's capabilities are restricted to text extraction and linking to other internal tools; it lacks the ability to execute arbitrary commands or perform external network requests directly.
Sanitization: There is no explicit sanitization or filtering of the input text to remove potential instructions hidden within legal documents.

struktura-wyroku-pl