MCP server for LeakIX API - security research and reconnaissance
Server has 10 tools with complete input schemas and descriptions. Naming follows verb_noun pattern (domain_lookup, host_lookup, search_leaks). Descriptions are adequate (100-200 chars) but lack specificity on when to use each tool vs. alternatives. Parameters are typed and described. Critical gap: no output schemas documented, LLMs cannot plan downstream calls or extract structured data. No error handling guidance. No tool annotations (readOnlyHint present via Risk field but not in MCP schema). Composition is sound (each tool does one thing), but missing pagination/limit parameters on search tools that could return large result sets.
Bulk export leak data (requires Pro API). Returns aggregated results for large-scale analysis. Use this for exporting large datasets efficiently. Results include grouped events by target.
Get information about a specific domain. Returns services and data leaks associated with the domain and its subdomains, including exposed services, certificates, and potential security issues.
Generate a comprehensive security exposure report. Analyzes a target and returns: risk level, critical findings, exposed ports, technologies, leak summary, and recommendations. Perfect for security assessments.
Find targets related to a given IP or domain. Discovers similar targets based on shared characteristics like technology stack, ASN, or network range. Useful for attack surface mapping.
Get information about a specific IP address. Returns services, leaks, and security data associated with the IP, including open ports, running services, and potential vulnerabilities.
No output schemas documented. LLMs cannot infer what fields each tool returns, forcing them to guess at response structure and breaking tool chaining. E.g., domain_lookup returns 'services and data leaks' but no schema shows field names, types, or pagination.
Search tools (search_leaks, search_services, bulk_export) lack pagination parameters (limit, offset, page_size). Large result sets will blow context windows. No mention of result caps or pagination strategy in descriptions.
No error handling guidance. Descriptions do not explain what happens on invalid input, API failures, or rate limits. LLMs have no recovery path if a call fails.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 62 | 2026-07-28+ | v2 |
List all available LeakIX plugins. Returns information about data sources and plugins used for reconnaissance.
Enumerate discovered subdomains for a domain. Returns a list of subdomains found through various discovery methods.
Quick reconnaissance on a target. Performs a fast scan combining multiple data sources to give a quick overview of a target's exposure.
Search for data leaks. Searches the LeakIX database for leaked data matching your query.
Search for exposed services. Searches for services matching your query across the LeakIX database.
Tool annotations missing from MCP schema. Risk field (READ_ONLY) is metadata but not exposed as MCP readOnlyHint. LLMs cannot infer tool safety from the protocol.
Descriptions lack 'when to use' guidance. domain_lookup vs host_lookup distinction is unclear (domain vs IP), and no description explains when to call find_related vs search_services. LLMs will guess wrong.