The Agent Skills Directory

[SAFE]: The skill implements a robust framework for extracting structured event data from legal materials. It explicitly instructs the agent to maintain high data integrity by distinguishing between event dates and document dates, utilizing confidence bands (high, medium, low), and documenting source gaps rather than filling them with inferred information.
[SAFE]: No evidence of malicious behavior was found. The skill does not attempt to exfiltrate data, access sensitive local files, or execute remote code. It uses standard methods (browser, web search, MCP, API) for data ingestion as appropriate for its intended legal research and litigation support functions.
[SAFE]: The skill provides clear guidance on handling sensitive material, instructing users to flag items for 'privilege_review' and ensuring that disputed events are not converted into advocacy prose without explicit labeling, which reduces the risk of misleading outputs.

building-chronologies