mcpField
O
Operant Mcp

Operant Mcp

Local
github.com/operantlabsSecurity26Updated 30w ago

About Operant Mcp

Operant MCP is a security testing MCP server with 51 tools across 19 modules for penetration testing, network forensics, memory analysis, and vulnerability assessment. It targets security researchers and penetration testers, and includes 8 methodology prompts to structure assessments.

Key features

  1. 0151 tools across 19 modules
  2. 02Web vulnerability tests for SQLi, XSS, SSRF and more
  3. 03PCAP network forensics with TLS and DNS analysis
  4. 04Volatility memory analysis and Linux rootkit detection
  5. 05CloudTrail log analysis and anomaly detection
  6. 068 methodology prompts

Use cases

  1. 01Running structured penetration tests against web applications
  2. 02Analyzing PCAP files for leaked credentials and port scans
  3. 03Examining memory dumps and malicious Office documents

MCP support

This project is an MCP server that exposes security testing tools and methodology prompts to MCP clients.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"operant": {
"command": "npx",
"args": [
"-y",
"operant-mcp"
]
}
}
}
Run commandnpx -y operant-mcp

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmoperant-mcp1.0.1

Tools

62

Saved tool list · updated 3w ago

ParameterTypeRequiredDescription
urlstringYesFull URL with query parameter, e.g. https://target/filter?category=Gifts
parameterstringYesVulnerable query parameter name, e.g. 'category'
valuestringYesLegitimate parameter value to base the injection on, e.g. 'Gifts'

Details

From the project README

operant-mcp

NPM Version npm downloads GitHub stars License: MIT TypeScript MCP Server smithery badge

Install in VS Code Install in Cursor

operant-mcp MCP server

Security testing MCP server with 51 tools for penetration testing, network forensics, memory analysis, and vulnerability assessment.

Quick Start

npx operant-mcp

Or install globally:

npm install -g operant-mcp
operant-mcp

Usage with Claude Code

Add to your MCP config:

{
  "mcpServers": {
    "operant": {
      "command": "npx",
      "args": ["-y", "operant-mcp"]
    }
  }
}

Tools (51)

SQL Injection (6)
  • sqli_where_bypass — Test OR-based WHERE clause bypass
  • sqli_login_bypass — Test login form SQL injection
  • sqli_union_extract — UNION-based data extraction
  • sqli_blind_boolean — Boolean-based blind SQLi
  • sqli_blind_time — Time-based blind SQLi
  • sqli_file_read — Read files via LOAD_FILE()
XSS (2)
  • xss_reflected_test — Test reflected XSS with 10 payloads
  • xss_payload_generate — Generate context-aware XSS payloads
Command Injection (2)
  • cmdi_test — Test OS command injection
  • cmdi_blind_detect — Blind command injection via sleep timing
Path Traversal (1)
  • path_traversal_test — Test directory traversal with encoding variants
SSRF (2)
  • ssrf_test — Test SSRF with localhost bypass variants
  • ssrf_cloud_metadata — Test cloud metadata access via SSRF
PCAP/Network Forensics (8)
  • pcap_overview — Protocol hierarchy and endpoint stats
  • pcap_extract_credentials — Extract FTP/HTTP/SMTP credentials
  • pcap_dns_analysis — DNS query analysis
  • pcap_http_objects — Export HTTP objects
  • pcap_detect_scan — Detect port scanning
  • pcap_follow_stream — Follow TCP/UDP streams
  • pcap_tls_analysis — TLS/SNI analysis
  • pcap_llmnr_ntlm — Detect LLMNR/NTLM attacks
Reconnaissance (7)
  • recon_quick — Quick recon (robots.txt, headers, common dirs)
  • recon_dns — Full DNS enumeration
  • recon_vhost — Virtual host discovery
  • recon_tls_sans — Extract SANs from TLS certificates
  • recon_directory_bruteforce — Directory brute-force
  • recon_git_secrets — Search git repos for secrets
  • recon_s3_bucket — Test S3 bucket permissions
Memory Forensics (3)
  • volatility_linux — Linux memory analysis (Volatility 2)
  • volatility_windows — Windows memory analysis (Volatility 3)
  • memory_detect_rootkit — Linux rootkit detection
Malware Analysis (2)
  • maldoc_analyze — Full OLE document analysis pipeline
  • maldoc_extract_macros — Extract VBA macros
Cloud Security (2)
  • cloudtrail_analyze — CloudTrail log analysis
  • cloudtrail_find_anomalies — Detect anomalous CloudTrail events
Authentication (3)
  • auth_csrf_extract — Extract CSRF tokens
  • auth_bruteforce — Username enumeration + credential brute-force
  • auth_cookie_tamper — Cookie tampering test
Access Control (2)
  • idor_test — Test for IDOR vulnerabilities
  • role_escalation_test — Test privilege escalation
Business Logic (2)
  • price_manipulation_test — Test price/quantity manipulation
  • coupon_abuse_test — Test coupon stacking/reuse
Clickjacking (2)
  • clickjacking_test — Test X-Frame-Options/CSP
  • frame_buster_bypass — Test frame-busting bypass
CORS (1)
  • cors_test — Test CORS misconfigurations
File Upload (1)
  • file_upload_test — Test file upload bypasses
NoSQL Injection (2)
  • nosqli_auth_bypass — MongoDB auth bypass
  • nosqli_detect — NoSQL injection detection
Deserialization (1)
  • deserialization_test — Test insecure deserialization
GraphQL (2)
  • graphql_introspect — Full schema introspection
  • graphql_find_hidden — Discover hidden fields

Prompts (8)

Methodology guides for structured security assessments:

  • web_app_pentest — Full web app pentest methodology
  • pcap_forensics — PCAP analysis workflow
  • memory_forensics — Memory dump analysis (Linux/Windows)
  • recon_methodology — Reconnaissance checklist
  • malware_analysis — Malware document analysis
  • cloud_security_audit — CloudTrail analysis workflow
  • sqli_methodology — SQL injection testing guide
  • xss_methodology — XSS testing guide

System Requirements

Tools require various CLI utilities depending on the module:

  • Most tools: curl
  • PCAP analysis: tshark (Wireshark CLI)
  • DNS recon: dig, host
  • Memory forensics: volatility / vol.py / vol3
  • Malware analysis: olevba, oledump.py
  • Cloud analysis: jq
  • Secrets scanning: git

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.