Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
maco144 avatar

Nullcone Threat Intelligence

  • Updated July 8, 2026
  • maco144/nullcone-mcp

Nullcone is a MCP server that delivers real-time threat intelligence with 890K+ IOCs for AI agents, including prompt-injection and skill threats.

About

Nullcone is a hosted threat-intelligence MCP server aimed at developers shipping AI agents and skills who need fast indicator lookups without standing up a full TI platform. The catalog positions it around a large IOC corpus—on the order of hundreds of thousands of entries—with emphasis on prompt-injection and AI-skill abuse patterns that generic blocklists miss. developers wire the streamable HTTP remote into Claude Code or Cursor during Ship security review: validate URLs, payloads, and skill supply chains before exposing automation to customers. It complements code review and dependency scanning rather than replacing them. Because it is a remote intel feed, you should still define retention, logging, and escalation policies in your own app; Nullcone supplies the enrichment layer agents can query at decision time.

  • Nullcone Threat Intelligence remote MCP at https://nullcone.ai/mcp
  • Catalog cites 890K+ IOCs including prompt-injection and AI-skill threats
  • streamable-http transport (v0.1.0) with website nullcone.ai
  • GitHub source maco144/nullcone-mcp for implementation reference
  • Focused on AI agent threat context, not generic enterprise SIEM dashboards

Nullcone Threat Intelligence by the numbers

  • Data as of Jul 9, 2026 (Skillselion catalog sync)
terminal
claude mcp add --transport http nullcone https://nullcone.ai/mcp

Add your badge

Show developers this MCP server is listed on Skillselion. Paste this into your README.

Listed on Skillselion
TransportHTTP
AuthNone
Last updatedJuly 8, 2026
Repositorymaco144/nullcone-mcp

What it does

Feed agents real-time IOC and AI-specific threat intelligence—including prompt-injection signals—before you ship agent features to users.

Who is it for?

Best when you're hardening Claude/Cursor agents and custom skills and want a zero-install remote TI MCP endpoint.

Skip if: Skip if you need on-prem-only intel with air-gapped policies or full SOC case management in one tool.

What you get

After adding the remote MCP, agents can query Nullcone for relevant IOCs and AI-centric threats during pre-ship security checks.

  • Live IOC and AI-threat queries via hosted MCP
  • Access to catalog-described 890K+ indicator corpus scope
  • Prompt-injection and AI-skill-oriented intelligence context for agents

By the numbers

  • Publisher cites 890K+ IOCs in the server description
  • Remote endpoint https://nullcone.ai/mcp (streamable-http)
  • Server manifest version 0.1.0; repository nullcone-mcp on GitHub
README.md

Nullcone MCP Server

Real-time threat intelligence for AI agents, exposed as a Model Context Protocol server. Check IPs, domains, URLs, hashes, CVEs, prompt-injection payloads, and malicious AI-skill / MCP-tool definitions against the Nullcone network — 890K+ IOCs, free, no API key.

Backed by nullcone.ai.

Use it (hosted — nothing to install)

The server is hosted at https://nullcone.ai/mcp over streamable HTTP. Add it to any MCP client:

Claude Code

claude mcp add --transport http nullcone https://nullcone.ai/mcp

Cursor / other MCP clients — add to your MCP config:

{
  "mcpServers": {
    "nullcone": {
      "url": "https://nullcone.ai/mcp"
    }
  }
}

No signup or token required. Read tools and IOC submission are open; destructive tools (e.g. revoke_ioc) are disabled on the public endpoint.

Tools

30+ tools including:

  • lookup_ioc(value) — check any indicator against the feed
  • recent_threats(limit, min_severity) — current threat picture
  • submit_ioc(...) / submit_batch(...) — contribute indicators
  • check_prompt(...) — sub-millisecond prompt-injection lookup
  • validate_skill(...) / scan_skill_content(...) — vet MCP tools / AI skills before loading
  • poll_since(last_id) — incremental sync, no persistent connection
  • get_stats(), list_families(), search_by_type(...), and more

Resources: threat://stats, threat://recent, threat://families, threat://family/{name}, threat://ioc/{value}. Prompts: analyze_ioc, triage_alert, threat_brief.

Self-host

The server is built on the public nullcone SDK.

pip install -r requirements.txt
MCP_TRANSPORT=streamable-http MCP_PORT=8001 python server.py

Or with Docker:

docker build -t nullcone-mcp .
docker run -p 8001:8001 nullcone-mcp

Set MCP_PUBLIC=1 to run an anonymous public endpoint (disables destructive tools); omit it for full local control over stdio (MCP_TRANSPORT=stdio).

License

Rising Sun License v1.0 — see LICENSE. Free for individuals and small teams.

Recommended MCP Servers

How it compares

AI-focused threat-intel MCP feed, not a code-review skill or dependency audit marketplace.

FAQ

Who is nullcone for?

and developers shipping AI agents or skills who need prompt-injection and IOC context inside their MCP toolchain.

When should I use nullcone?

Use it in Ship security passes when validating agent inputs, skill sources, and external payloads against live threat data.

How do I add nullcone to my agent?

Register the remote MCP URL https://nullcone.ai/mcp (streamable-http) in your client; no local package is listed in the server manifest.

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.