官方说明📇 ☁️ - MCP server for querying the ORKL API. This server provides tools for fetching threat reports, analyzing threat actors, and retrieving intelligence sources.
配置模板(请按你的客户端调整)
claude mcp add mcp-security -- npx -y mcp_security
A full write-up can be found here: https://blog.securitybreak.io/building-a-threat-intelligence-genai-reporter-with-orkl-and-claude-a0ae2e969693 A Model Context Protocol (MCP) server for querying the ORKL API. This server provides tools for fetching and analyzing threat reports, threat actors, and sources. It integrates smoothly with MCP-compatible applications. --- Quick Install Edit or create the file /Users/user/Library/Application Support/Claude/claude_desktop_config.json --- Tools Report Tools Fetch Latest Threat Reports - **Name**: fetch_latest_threat_reports - **Description**: Fetch recent threat reports with their titles and IDs. - **Parameters**: None Fetch Threat Report Details - **Name**: fetch_threat_report_details - **Description**: Retrieve detailed information for a specific threat report by ID. - **Parameters**: - report_id (required): The ID of the threat report. Threat Actor Tools Fetch Threat Actors - **Name**: fetch_threat_actors - **Description**: Fetch a list of known threat actors with their IDs and names. - **Parameters**: None Fetch Threat Actor Details - **Name**: fetch_threat_actor_details - **Description**: Retrieve detailed information for a specific threat actor by ID. - **Parameters**: - actor_id (required): The ID of the threat actor. Source Tools Fetch Sources - **Name**: fetch_sources - **Description**: Fetch a list of sources used in threat intelligence. - **Parameters**: None Fetch Source Details - **Name**: fetch_source_details - **Description**: Retrieve detailed metadata for a specific source by ID. - **Parameters**: - source_id (required): The ID of the source.