The Agent Skills Directory

[SAFE]: The skill serves as a utility for converting legal research and case analysis into a structured JSON format. It explicitly directs the agent to prioritize structured data sources and validated citations, which reduces the risk of hallucination and data corruption.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes external legal texts (facts and contracts) and writes them to a project file.
Ingestion points: Case facts, legal summaries, and contract text provided by users or prior analysis steps in the agent context.
Boundary markers: No explicit delimiters or instructions are used to separate user-provided data from the agent's logic during the JSON assembly process.
Capability inventory: The skill requires file-writing capabilities to update the project file law-powers/data.js.
Sanitization: The instructions do not specify sanitization or escaping protocols for content stored in the description or label fields of the output JSON.

legal-graph