A production-ready Model Context Protocol server that exposes all 19 LeadMagic API endpoints as MCP tools. Provides seamless integration with MCP-compatible clients like Claude Desktop, Cursor, Windsurf, and others for B2B data enrichment, email finding, and company intelligence.
The LeadMagic MCP server exposes 17 tools with solid naming conventions (verb_noun pattern consistently applied) and reasonable descriptions. However, there are significant gaps in parameter documentation, output schema definition, and error handling guidance. Tools are well-named and descriptions are present but lack the depth required for production-grade LLM reasoning. Parameters have type definitions but many lack detailed descriptions explaining constraints, formats, or dependencies. No output schemas are documented, making it impossible for LLMs to plan downstream tool calls. Error handling is minimal, no recovery guidance, no categorization (retryable vs fatal), and no actionable messages. The server is functional but falls short of patterns for tool composition, response shaping, and LLM-optimized documentation.
Find verified email address based on a person's name and company. Combines multiple data sources to locate professional email addresses with high accuracy.
Find employees at a specific company. Retrieves a list of employees from target company with their profiles, roles, and contact information.
Search for job openings and recruitment data. Finds active job postings with detailed company information, location data, and job specifics across multiple job boards.
Find mobile phone numbers for professional contacts. Locates verified personal and business phone numbers with geographic and carrier information.
Find personal email addresses from B2B profile URLs. Useful for reaching contacts through their personal channels when professional emails are not available.
No output schemas documented for any tool. LLMs cannot plan downstream tool calls or know what fields to extract from responses.
Parameter descriptions are minimal or missing detailed constraints. For example, 'find_email' has parameters 'domain' and 'company_name' marked as alternatives but no description explaining mutual exclusivity, required format, or what constitutes a valid domain. Rubric requires: 'When parameters are mutually exclusive (e.g. user_id vs email), state this in descriptions.'
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | C | 63 | 2026-07-28+ | v2 |
Find people in specific roles at target companies. Locates professionals with specific job titles, functions, or roles for recruitment and sales intelligence.
Get detailed information about a specific advertisement. Retrieves complete details about a B2B advertisement including creative, targeting, and performance data.
Get company funding and investment information. Retrieves detailed funding rounds, investors, valuations, and financial history for target companies.
Check the number of available API credits for your LeadMagic account. Essential for monitoring usage and planning API calls.
Search B2B advertisement data and advertising intelligence. Retrieves B2B-specific advertising data for competitive analysis and market intelligence.
Search for complete B2B profile information from URL. Retrieves comprehensive profile data including personal, professional, and company information from B2B profile URLs.
Search for detailed company information using domain, name, or profile URL. Provides comprehensive business intelligence including employee count, locations, industry details, and company metadata.
Search Google advertisement data for target companies. Retrieves information about Google Ads campaigns, keywords, and advertising activity for competitive intelligence.
Search Meta (Facebook/Instagram) advertisement data. Retrieves information about Meta Ads campaigns and advertising activity for target companies and accounts.
Get full profile details from B2B profile URL (e.g., LinkedIn). Extracts comprehensive professional information including work history, education, and company details. Rate limit: 300 requests/minute.
Find work email addresses from B2B profile URLs. Converts social profile information into professional email contacts.
Validate an email address for deliverability and retrieve associated company information. Provides detailed analysis of email validity, MX records, and catch-all detection.
No pagination guidance. Tools like 'find_employees', 'find_jobs', 'find_role' accept a 'limit' parameter but provide no page/offset, no cursor, no total count in response. Rubric requires: 'Tools returning lists should accept page/offset and limit parameters and return a total count or next_cursor. Without pagination, large results blow the context window.'
Error handling is absent from visible code. No recovery guidance, no error categorization (retryable vs fatal), no actionable error messages. Rubric requires: 'Error responses must tell the LLM what to do next' and 'Categorize errors as retryable, user-fixable, or fatal.'
Parameter descriptions lack format constraints. 'profile_url' parameters (find_personal_email, social_to_work_email, search_profile, search_b2b_profile) have no guidance on expected URL format, platform (LinkedIn only?), or what constitutes a valid B2B profile URL. LLMs will guess and pass malformed inputs.
Rate limiting mentioned in description for 'search_profile' ('Rate limit: 300 requests/minute') but not enforced or documented for other tools. No guidance on what happens when rate limit is exceeded, how to backoff, or whether retries are safe.
No response field naming consistency guidance. If 'find_email' returns an 'email' field, but 'validate_email' returns 'email_address', LLMs will struggle to chain outputs to inputs. Rubric requires: 'Mismatched naming forces the LLM to reason about field mappings, increasing errors.'
Similar tools lack clear differentiation. 'find_email' vs 'validate_email' vs 'search_profile' vs 'search_b2b_profile' are semantically close. Descriptions do not clearly distinguish when to use one over another, forcing LLM reasoning cycles and potential misuse.