FAQ
Frequently asked questions about FullHunt Agentic AI.
General Questions
What is FullHunt Agentic AI?
FullHunt Agentic AI is an AI-powered interface to FullHunt's attack surface intelligence platform, built on Anthropic's Model Context Protocol (MCP). It enables AI assistants like Claude to autonomously conduct security reconnaissance, vulnerability research, and threat intelligence operations through natural language.
How does it differ from the REST API?
Feature | REST API | Agentic AI (MCP) |
---|---|---|
Interface | HTTP requests | Natural language |
Complexity | Requires coding | No coding required |
Workflow | Manual chaining | Autonomous chaining |
Learning Curve | API documentation | Conversational |
Use Case | Automation scripts | Interactive analysis |
Both interfaces access the same underlying data and capabilities.
Which accounts support Agentic AI?
- ✅ Professional accounts - Full access to public API tools
- ✅ Enterprise accounts - Full access including enterprise-only tools
- ❌ Free accounts - Not currently supported
Upgrade your account to access Agentic AI.
What AI assistants are supported?
FullHunt Agentic AI works with any MCP-compatible client:
- ✅ Claude Desktop (recommended)
- ✅ Cursor IDE
- ✅ Custom MCP clients
- ✅ Any client supporting HTTP transport MCP
Is the MCP server open source?
No, FullHunt MCP is a hosted service. You connect to our endpoints:
- Public API:
https://fullhunt.io/api/v1/mcp
- Enterprise API:
https://enterprise-api.fullhunt.io/api/v1/mcp
No local server installation required!
Technical Questions
How do I get started?
- Get a FullHunt API key from Profile Settings
- Install Claude Desktop or another MCP client
- Add FullHunt configuration to your MCP client
- Start asking security questions in natural language!
See the Integration Guide for detailed setup.
Where do I get my API key?
Get your API key from your FullHunt Profile Settings.
- Log in to FullHunt
- Go to User Settings
- Navigate to API Keys section
- Copy your API key or generate a new one
How do I configure Claude Desktop?
Add this to your Claude Desktop config:
macOS/Linux: ~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"fullhunt": {
"transport": "http",
"url": "https://fullhunt.io/api/v1/mcp",
"headers": {
"X-API-KEY": "your-api-key-here"
}
}
}
}
See Integration Guide for complete setup.
Can I use environment variables for API keys?
Yes! This is the recommended approach:
{
"mcpServers": {
"fullhunt": {
"transport": "http",
"url": "https://fullhunt.io/api/v1/mcp",
"headers": {
"X-API-KEY": "${env:FULLHUNT_API_KEY}"
}
}
}
}
Then set the environment variable:
export FULLHUNT_API_KEY="your-api-key-here"
What endpoints should I use?
Professional/Standard accounts:
https://fullhunt.io/api/v1/mcp
Enterprise accounts:
https://enterprise-api.fullhunt.io/api/v1/mcp
How many tools are available?
Over 40 production tools covering:
- Domain & Host Intelligence (6 tools)
- Vulnerability Intelligence (2 tools)
- Attack Surface Management (2 tools)
- Data Intelligence (12 tools)
- Enterprise APIs (12 tools)
- Nexus Intelligence (6 tools)
- OEM APIs (5 tools)
See Tools Reference for complete list.
Usage Questions
What can I ask the AI?
You can ask security-related questions in natural language:
Examples:
- "Find all subdomains of example.com"
- "Search for CVE-2024-1234 and check if we're vulnerable"
- "Is IP 1.2.3.4 a Tor exit node?"
- "Find exposed admin panels on example.com"
- "Search dark web for compromised credentials"
See Examples for more query ideas.
Can the AI chain multiple operations?
Yes! The AI can autonomously perform multi-step investigations:
"Investigate acme.com security:
1. Find all subdomains
2. Identify technologies used
3. Search for vulnerabilities
4. Check dark web for breaches
5. Generate security report"
The AI will execute each step and provide a comprehensive analysis.
How fast are queries?
Most queries complete in seconds:
- Simple lookups: 1-2 seconds
- Complex scans: 5-10 seconds
- Multi-step workflows: 10-30 seconds
Performance depends on query complexity and account tier.
What's the difference between tools?
Different tools serve different purposes:
- fullhunt_host: Real-time host scanning
- intel_host: Historical intelligence data
- nexus_ip_lookup: Threat intelligence
See Tools Reference for detailed explanations.
Security Questions
Is my API key secure?
Your API key security depends on how you manage it:
✅ Do:
- Store in environment variables
- Use separate keys for different environments
- Rotate keys regularly
- Revoke unused keys
❌ Don't:
- Commit to version control
- Share in chat/email
- Hardcode in scripts
See Best Practices - Security for details.
What data does FullHunt collect?
FullHunt collects:
- API queries and responses
- Usage statistics
- Error logs
- Performance metrics
We do NOT collect or store:
- Your API keys
- Conversation content outside API calls
- Personal data beyond account info
Can I test on any domain?
⚠️ Important: Only test domains you own or have explicit permission to test.
Allowed:
- Your own domains
- Domains with written authorization
- Public information research
Not allowed:
- Unauthorized third-party testing
- Malicious reconnaissance
- Violating terms of service
Is the traffic encrypted?
Yes, all traffic is encrypted:
- HTTPS/TLS for all API communication
- API keys transmitted via secure headers
- Data at rest encrypted in FullHunt infrastructure
Enterprise Questions
What extra features do Enterprise accounts get?
Enterprise accounts include:
- Dedicated MCP endpoint
- Organization management tools
- Real-time security alerts
- Vulnerability tracking
- Dark web monitoring
- Custom entity management
- Priority support
See Enterprise APIs for details.
How do I get an Enterprise account?
Contact our team at team@fullhunt.io to discuss:
- Your organization's needs
- Custom pricing
- Dedicated infrastructure
- Enterprise features
Can I get a dedicated tenant?
Yes! Enterprise customers can get:
- Dedicated API infrastructure
- Custom domain endpoints
- Isolated data storage
- Enhanced SLAs
Contact team@fullhunt.io for enterprise options.
Troubleshooting
"Failed to connect to MCP server"
Solutions:
- Check your internet connection
- Verify the endpoint URL is correct
- Ensure your API key is valid
- Restart your MCP client
"Authentication failed"
Solutions:
- Verify your API key is correct (no extra spaces)
- Check your account is active
- Regenerate API key if compromised
- Ensure you're using the right endpoint
"Tool not found" or "Access denied"
Solutions:
- Verify your account supports the tool
- Check tool name spelling
- Enterprise tools require Enterprise account
- Review Tools Reference
The AI isn't using FullHunt tools
Solutions:
- Verify MCP configuration is loaded
- Restart your AI client
- Check configuration file location
- Ensure API key is valid
- Try explicitly mentioning FullHunt
Slow responses
Solutions:
- Break complex queries into smaller parts
- Use more specific searches
- Verify network connection
See Integration Guide - Troubleshooting for more.
Support
Where can I get help?
Documentation:
Support Channels:
- Email: team@fullhunt.io
- Professional Support: Included with Professional plan
- Dedicated Support: Included with Enterprise plan
How do I report a bug?
Email team@fullhunt.io with:
- Description of the issue
- Query that caused the problem
- Expected vs. actual behavior
- Your account email
- Timestamp of the issue
Can I request new features?
Yes! We welcome feature requests. Email team@fullhunt.io with:
- Feature description
- Use case
- Expected benefits
Is there a community?
Yes! Join the FullHunt community:
- Share workflows and tips
- Learn from other users
- Contribute to best practices
- Get updates on new features
More Questions?
Can't find your answer? Contact us:
- Email: team@fullhunt.io
- Documentation: api-docs.fullhunt.io
- Website: fullhunt.io
Ready to get started? Sign up for FullHunt today!