misp-mcp
ActiveMCP server for MISP - threat-intelligence sharing and lookups.
WHAT IT DOES
misp-mcp is a Model Context Protocol server for
MISP,
the open-source threat-intelligence sharing platform. It lets an LLM run IOC lookups,
create and manage events, discover cross-event correlations, and export indicators
directly from your MISP instance.
It exposes 36 tools, 3 resources, and 3 guided prompts spanning events, attributes, correlations, tags, exports, sightings, warninglists, objects, galaxies, feeds, organisations, and server management. MITRE ATT&CK is wired in through galaxy-cluster search and attachment, and a correlation engine surfaces relationships through shared indicators.
State-changing and destructive tools (delete, publish, tag-removal) are guarded: they
refuse to run without an explicit confirm: true, and permanent hard
deletes require a second confirmHard: true. SSL verification is flexible
for the self-signed certs common in MISP deployments.
INSTALL
Published to npm. Install globally to get the misp-mcp binary, or run it
on demand with npx. Requires Node.js 20+ and a MISP instance with API
access plus an API key (Administration > List Auth Keys).
Configure with MISP_URL and MISP_API_KEY. Set
MISP_VERIFY_SSL=false for self-signed certs.
CLIENT CONFIG
A sample mcpServers entry (Claude Desktop, or any MCP client that reads the
same shape). It runs the installed binary over stdio.
{
"mcpServers": {
"misp": {
"command": "misp-mcp",
"env": {
"MISP_URL": "https://misp.example.com",
"MISP_API_KEY": "your-api-key-here",
"MISP_VERIFY_SSL": "false"
}
}
}
}
For Claude Code use claude mcp add misp ... -- misp-mcp. The repo README
also documents OpenClaw, Codex CLI, Hermes Agent, and Docker.
CAPABILITIES / TOOLS EXPOSED
36 MCP tools across events, attributes, correlation, exports, tags, objects, galaxies, feeds, and server admin. Grouped below; see the README for the full per-argument reference.
Events
Attributes & correlation
Exports & sightings
Tags, objects & galaxies
Feeds, orgs & server
Plus 3 resources (misp://types, misp://statistics,
misp://taxonomies) and 3 prompts (investigate-ioc, create-incident-event,
threat-report).