MISP MCP Server
A Model Context Protocol (MCP) server that integrates with the MISP (Malware Information Sharing Platform) to provide threat intelligence capabilities to Large Language Models.
Features
- Mac Malware Detection: Search for the latest macOS-related malware samples
- Cross-Platform Threat Intelligence: Search for threats affecting Windows, macOS, Linux, Android, iOS, and IoT devices
- Advanced Search Capabilities: Search by attribute type, tag, threat actor, or TLP classification
- IoC Submission: Submit new Indicators of Compromise directly to your MISP instance
- Threat Intelligence Reports: Generate comprehensive reports based on MISP data
- MISP Statistics: Get insights into your MISP instance's data
Prerequisites
- Python 3.10 or higher
- MISP instance with API access
- API key with appropriate permissions
Installation
- Clone this repository:Copy
- Create a virtual environment and install dependencies:Copy
Configuration
Set the following environment variables to connect to your MISP instance:
MISP_URL
- URL of your MISP instance (e.g., "https://misp.example.com")MISP_API_KEY
- Your MISP API keyMISP_VERIFY_SSL
- Whether to verify SSL certificates (True/False)
Usage
Running as a standalone server
Testing with MCP Inspector
Installing in Claude Desktop
Edit your Claude Desktop configuration file:
macOS:
Windows:
Add the MISP MCP server configuration:
Alternatively, use the MCP CLI:
Available Tools
get_mac_malware
Get the latest Mac-related malware samples from MISP.
Parameters:
days
(default: 30): Number of days to look backlimit
(default: 10): Maximum number of results to return
get_platform_malware
Get the latest malware samples for a specific platform from MISP.
Parameters:
platform
: Platform to search for (windows, macos, linux, android, ios, iot)days
(default: 30): Number of days to look backlimit
(default: 10): Maximum number of results to return
advanced_search
Perform advanced searches in MISP.
Parameters:
query_type
: Type of search (attribute_type, tag, threatactor, tlp)query_value
: Value to search forplatform
(optional): Platform filter (windows, macos, linux, android, ios, iot)days
(default: 30): Number of days to look backlimit
(default: 10): Maximum number of results to return
submit_ioc
Submit a new Indicator of Compromise (IoC) to MISP.
Parameters:
ioc_value
: The actual IoC value (e.g., hash, URL, IP)ioc_type
: Type of IoC (e.g., md5, sha256, url, ip-dst, filename)event_info
: Brief description of the eventcategory
(default: "Artifacts dropped"): Category of the attributeplatform
(optional): Platform affected (windows, macos, linux, android, ios, iot)tlp
(default: "amber"): Traffic Light Protocol level (white, green, amber, red)comment
(optional): Optional comment for the IoC
generate_threat_report
Generate a comprehensive threat intelligence report based on MISP data.
Parameters:
days
(default: 30): Number of days to include in the reportplatforms
(default: "all"): Comma-separated list of platforms or "all"threat_level
(default: "all"): Filter by threat level (low, medium, high, all)include_stats
(default: True): Whether to include statistics
search_misp
Search MISP for specific threats.
Parameters:
query
: Search term (e.g., CVE ID, malware name, hash)days
(default: 30): Number of days to look back
get_misp_stats
Get statistics about the MISP instance.
Available Resources
feeds://recent/{days}
Get information about recent MISP feeds.
Parameters:
days
(default: 7): Number of days to look back
Example Queries with Claude
- "What are the latest Mac-related malware samples?"
- "Show me Windows malware from the last 2 weeks"
- "Search for CVE-2023-12345 in MISP"
- "Submit this IoC to MISP: 1a2b3c4d5e6f7g8h9i0j, type: md5, description: suspicious file found in phishing email"
- "Generate a threat intelligence report for the last month"
- "What are the current MISP statistics?"
- "Get information about recent MISP feeds"
- "Perform an advanced search for TLP:RED events related to banking trojans"
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
This project is licensed under the MIT License - see the LICENSE file for details.