mitre-attack-mcp
MITRE ATT&CK MCP Server
Description
A Model Context Protocol server that exposes the MITRE ATT&CK knowledge base to AI agents such as Claude Desktop, Claude Code and LangGraph. Use for Tactic, Technique, Procedure, Mitigation and Detection lookups. PyPi - https://pypi.org/project/mitre-attack-mcp, Githib - https://github.com/gsbadhan/mitre-attack-mcp .
What it does
| Tool | Purpose | Input |
|---|---|---|
get_data_stats |
Counts of techniques, groups, software, mitigations | optional domain |
search_techniques |
Find techniques by keyword in name or description | keyword; optional domain, limit (default 10) |
get_technique_by_id |
Technique details by ID (name, description, tactics, platforms) | attack_id; optional domain |
get_techniques_by_tactic |
Techniques belonging to a tactic (by ID like TA0006 or shortname like execution) |
tactic; optional domain |
get_procedures_for_technique |
Procedure examples: groups and software that used the technique, and how | attack_id; optional domain |
get_mitigations_for_technique |
Mitigations (M####) for a technique | attack_id; optional domain |
get_detections_for_technique |
Detection strategies (DET####) and platform-specific analytics (AN####) | attack_id; optional domain |
get_group_by_name |
Threat group details by name or alias | name; optional domain |
All tools also accept an optional domain: enterprise-attack, ics-attack, or mobile-attack.
Project setup
- uv venv
- source .venv/bin/activate
- uv sync
Data ingestion
- MITRE data auto downloaded while MCP server starts
Run MCP server locally
- uv run python -m mitre_attack_mcp.server
Note: Dont run with Claude, its for local testing only
Unit tests
- uv run pytest tests/ -v
Integration With Anthropic Claude Code
1. Clone code repository locally
2. Check mcps: claude mcp list
3. Add below configs in macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
"mcpServers": {
"mitre-attack": {
"command": "uv",
"args": [
"run",
"--directory",
"/Users/john/git-projects/mitre-attack-mcp",
"mitre-attack-mcp"
]
}
}
4. Register mitre mcp: claude mcp add --scope user mitre-attack -- uv run --directory /Users/john/git-projects/mitre-attack-mcp mitre-attack-mcp
5. Check mitre-attack mcp: claude mcp list
6. Check detail: claude mcp get mitre-attack
Integration with Anthropic Claude Desktop
1. Claude desktop -> Settings -> Developer -> Edit config
2. Add below configs if you using git repository code:
"mcpServers": {
"mitre-attack": {
"command": "uv",
"args": [
"run",
"--directory",
"/Users/john/git-projects/mitre-attack-mcp",
"mitre-attack-mcp"
]
}
}
2. Add below configs if you using PyPi:
"mcpServers": {
"mitre-attack": {
"command": "uvx",
"args": ["mitre-attack-mcp@latest"]
}
}
3. Relaunch Claude
Claude Desktop Assistant With Mitre MCP
Common prompts
Search mitre-attack for techniques containing 'credential'.
Get the data stats from mitre-attack.
Look up the group APT29 in mitre-attack.
Use the mitre-attack server to look up technique T1059.
what is mitigations for T1003.
show tactic, procedure, mitigations and detections of T1059 ?
Metadata
Release files for mitre-attack-mcp 0.1.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| mitre_attack_mcp-0.1.4.tar.gz | 448.5 kB | Without provenance | Details for mitre_attack_mcp-0.1.4.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| mitre_attack_mcp-0.1.4-py3-none-any.whl | Python 3 | none | any | Details for mitre_attack_mcp-0.1.4-py3-none-any.whl |
Total release size: 458.9 kB
Release files / mitre_attack_mcp-0.1.4.tar.gz
| Download URL | mitre_attack_mcp-0.1.4.tar.gz |
|---|---|
| Size | 448.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0885f273aef7dbfac43fd9b9412b7c25dd9f80f002cfd5f1740a87a1a4f777bf
|
|
BLAKE2b-256 checksum How to use checksums |
b0c2bb73da665f9cea03aa993e8a112a08e43d4219f58b559b3c5745adb3bb8d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.7.13
|
Release files / mitre_attack_mcp-0.1.4-py3-none-any.whl
| Download URL | mitre_attack_mcp-0.1.4-py3-none-any.whl |
|---|---|
| Size | 10.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
11199ff53cde289f509e4e5e82398dba8b61d2998f0c4065f8a42d776f36cd35
|
|
BLAKE2b-256 checksum How to use checksums |
5ad395836c70daeb8b6f7c4e6bf16ed6b9c87f21aa0a3b6811d40ac76f88e30f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.7.13
|
Security for mitre-attack-mcp 0.1.4
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.