Provides revenue intelligence tools (RFM analysis, ICP triangulation, pipeline health scoring) for HubSpot data, enabling AI agents to analyze and segment prospects and pipeline.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Artefact Revenue Intelligence MCP Server qualify TechStart Inc"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Three-dimensional revenue intelligence. Because demographics alone don't close deals.
A Model Context Protocol (MCP) server that gives AI agents access to the ICP Triangulation Framework™ — scoring prospects across firmographics, behaviors, and growth signals. Plus RFM analysis and pipeline health scoring. Built on the Artefact Formula methodology from real B2B consulting engagements.
Traditional ICP models stop at firmographics. We triangulate across three dimensions to identify prospects with the right profile, the right behaviors, AND the right trajectory.
Feature | HubSpot Official MCP | Generic Wrappers | Artefact MCP |
CRUD operations | Yes | Yes | Via HubSpot API |
RFM Analysis | No | No | 11-segment classification |
ICP Triangulation | No | No | Firmographic + Behavioral + Growth Signals |
Pipeline Health | No | No | 0-100 health score |
Methodology built-in | No | No | Artefact Formula |
Works without API key | No | No | Yes (demo data) |
Related MCP server: artefact-mcp-server
B2B revenue teams using HubSpot who want AI-powered customer segmentation
RevOps managers who need pipeline health analysis accessible from Claude or Cursor
Consultants who deliver RFM analysis and ICP scoring to clients
Developers building revenue intelligence integrations with MCP
run_rfm — RFM AnalysisScores clients on Recency, Frequency, and Monetary value. Segments them into 11 categories (Champions through Lost) and extracts ICP patterns from top performers. Supports B2B service, SaaS, and manufacturing presets.
qualify — ICP Triangulation Framework™Go beyond demographics. Scores prospects across three dimensions to identify revenue-ready opportunities:
🏢 Firmographic Fit (Who they are): Industry, revenue, employees, geography
🎯 Behavioral Fit (What they're doing): Tech stack, growth signals, engagement, purchase history
📈 Growth Signals (Where they're heading): Hiring, funding, expansion momentum
Returns tier classification (Ideal / Strong / Moderate / Poor) with engagement strategy. Technical implementation: 14.5-point scoring model.
score_pipeline_health — Pipeline Health ScoreAnalyzes open deals for velocity metrics, stage-to-stage conversion rates, bottleneck identification, and at-risk deal detection. Returns a 0-100 health score.
URI | Description |
| ICP Triangulation Framework technical reference |
| 4-tier classification system |
| 11 RFM segment definitions with scoring scales |
| SPICED discovery framework |
| HubSpot data setup and enrichment requirements for ICP triangulation |
⚠️ Important: The qualify tool requires specific data across all three dimensions:
✅ Native HubSpot data (Firmographic + Partial Behavioral):
Firmographic Fit: Industry, revenue, employees, geography — standard properties
Behavioral Fit (Partial): Tech stack, content engagement, purchase history — custom properties or workflows
⚠️ Requires external enrichment (Clay, Clearbit, or manual research):
Growth Signals (Behavioral Fit — Critical Dimension): Hiring trends, funding rounds, product launches, expansion signals, press mentions
HubSpot does NOT track growth signals natively
Without growth signals: You lose the third dimension of triangulation — prospect momentum and buying power indicators
See full guide: Ask your AI assistant to read methodology://data-requirements for complete setup instructions and Clay integration workflow.
pip install artefact-mcpnpx @smithery/cli install artefact-revenue-intelligenceclaude mcp add artefact-revenue -- uvx artefact-mcpThen ask:
"Run an RFM analysis on our HubSpot data"
"Qualify this prospect: SaaS company, $5M revenue, 80 employees in Ontario"
"Score our pipeline health"
Add to claude_desktop_config.json:
Recommended (Python method):
{
"mcpServers": {
"artefact-revenue": {
"command": "python3",
"args": ["-m", "artefact_mcp"],
"env": {
"HUBSPOT_API_KEY": "pat-na1-xxxxxxxx"
}
}
}
}Alternative (uvx method):
{
"mcpServers": {
"artefact-revenue": {
"command": "uvx",
"args": ["artefact-mcp"],
"env": {
"HUBSPOT_API_KEY": "pat-na1-xxxxxxxx"
}
}
}
}Note: If using uvx and seeing "Server disconnected" errors, see the Troubleshooting section below.
Add to .cursor/mcp.json:
Recommended (Python method):
{
"mcpServers": {
"artefact-revenue": {
"command": "python3",
"args": ["-m", "artefact_mcp"],
"env": {
"HUBSPOT_API_KEY": "pat-na1-xxxxxxxx"
}
}
}
}Alternative (uvx method):
{
"mcpServers": {
"artefact-revenue": {
"command": "uvx",
"args": ["artefact-mcp"],
"env": {
"HUBSPOT_API_KEY": "pat-na1-xxxxxxxx"
}
}
}
}from artefact_mcp.tools.rfm import run_rfm_analysis
from artefact_mcp.tools.icp import qualify_prospect
from artefact_mcp.tools.pipeline import score_pipeline
# RFM with sample data (no HubSpot key needed)
results = run_rfm_analysis(source="sample", industry_preset="b2b_service")
# ICP qualification
score = qualify_prospect(company_data={
"industry": "SaaS",
"annual_revenue": 10_000_000,
"employee_count": 80,
"geography": "Quebec",
"tech_stack": ["HubSpot", "Google Analytics"],
"growth_signals": ["hiring", "funding"],
"content_engagement": "active",
"decision_maker_access": "c_suite",
"budget_authority": "dedicated",
"strategic_alignment": "strong",
})
# Pipeline health
health = score_pipeline(source="sample")Problem: Claude Desktop shows "MCP artefact-revenue: Server disconnected" error when using uvx as the command.
Cause: Claude Desktop (and other sandboxed applications) may not have access to uvx in your PATH. This commonly happens when uvx is installed via:
Homebrew → ~/.local/bin/uvx
curl installation → ~/.cargo/bin/uvx or other locations
Solutions:
Use Python method (recommended): Switch to python3 -m artefact_mcp method (see Claude Desktop section above). Python is always in PATH.
Use full uvx path: Find your uvx location and use the full path:
# Find uvx location
which uvx
# Example output: /Users/yourname/.local/bin/uvxThen update your config with the full path:
{
"mcpServers": {
"artefact-revenue": {
"command": "/Users/yourname/.local/bin/uvx",
"args": ["artefact-mcp"],
"env": {}
}
}
}Verify manually: Test that the MCP server starts correctly:
uvx artefact-mcp==0.2.3
# Should see: "Artefact Revenue Intelligence MCP Server running..."Issue: Tools return "No HubSpot API key" errors.
Solution: Ensure HUBSPOT_API_KEY is set in your MCP server configuration. Or use source="sample" to test with demo data first.
Issue: Import errors when using python3 -m artefact_mcp.
Solution: Ensure the package is installed: pip install artefact-mcp or pip install --upgrade artefact-mcp.
Variable | Required | Description |
| No | HubSpot private app token. Without it, tools work with |
| No | License key for Pro/Enterprise tier. Free tier (sample data) works without a key. |
| No | Path to JSON file with custom HubSpot property mappings (Pro/Enterprise only). |
| No | Path to JSON file with custom RFM scoring thresholds (Pro/Enterprise only). |
If your HubSpot instance uses custom property names for behavioral and strategic fit data, you can configure property mappings. This allows the qualify tool to automatically fetch and score all ICP dimensions from your HubSpot data.
Create a JSON configuration file (e.g., artefact_property_mapping.json):
{
"tech_stack": "technologies_used",
"tech_stack_delimiter": ",",
"growth_signals": ["linkedin_hiring_count", "recent_funding_amount", "press_mentions"],
"growth_signal_keywords": {
"linkedin_hiring_count": "hiring",
"recent_funding_amount": "funding",
"press_mentions": "press"
},
"content_engagement": "hubspot_engagement_score",
"content_engagement_thresholds": {
"active": 10,
"occasional": 3
},
"decision_maker_access": "primary_contact_role",
"budget_authority": "budget_category",
"strategic_alignment": "revenue_ops_conviction"
}Set the environment variable:
export ARTEFACT_PROPERTY_MAPPING_PATH=/path/to/artefact_property_mapping.jsonAvailable Configuration Options:
Property | Type | Description | Default |
| string | HubSpot property name for tech stack | None |
| string | Delimiter for parsing text fields |
|
| array | List of HubSpot properties indicating growth | None |
| object | Map property names to signal keywords |
|
| string | HubSpot property for engagement score | None |
| object | Thresholds for active/occasional |
|
| string | Strategic fit property | None |
| string | Budget authority property | None |
| string | Strategic alignment property | None |
Example HubSpot Properties:
Common custom properties to map:
Tech Stack: tech_stack_used, technologies, crm_platform
Growth Signals: linkedin_job_postings_count, recent_funding_round, press_mentions_count, new_office_opened
Content Engagement: hs_analytics_num_page_views, email_engagement_score
Strategic Fit: primary_contact_role, budget_category, growth_conviction
The qualify tool will automatically fetch and score these custom properties when a property mapping is configured.
Example Configuration Files:
Two example configurations are included in the repository:
property_mapping.example.json — Full configuration with all available options
property_mapping.minimal.example.json — Minimal configuration for growth signals only
Copy the appropriate example file and customize it for your HubSpot instance:
cp property_mapping.minimal.example.json my_property_mapping.json
# Edit my_property_mapping.json with your HubSpot property names
export ARTEFACT_PROPERTY_MAPPING_PATH=$(pwd)/my_property_mapping.jsonPro/Enterprise users can customize RFM scoring thresholds to match their industry or business model. The built-in presets (b2b_service, saas, manufacturing) may not perfectly fit your buying cycles or revenue ranges.
Create an RFM threshold configuration file (e.g., rfm_thresholds.json):
{
"recency_days": [60, 180, 365, 730],
"recency_scores": [5, 4, 3, 2, 1],
"frequency_counts": [5, 3, 2, 1],
"frequency_scores": [5, 4, 3, 2, 1],
"monetary_method": "percentile",
"monetary_percentiles": [80, 60, 40, 20]
}Set the environment variable:
export ARTEFACT_RFM_THRESHOLDS_PATH=/path/to/rfm_thresholds.jsonAvailable Configuration Options:
Property | Type | Description | Default |
| array | Days since last purchase thresholds |
|
| array | Scores for each recency band (5 = best) |
|
| array | Transaction count thresholds |
|
| array | Scores for each frequency band |
|
| string | Scoring method: |
|
| array | Percentile thresholds (for percentile method) |
|
| array | Fixed dollar thresholds (for fixed method) |
|
| array | Scores for each monetary band |
|
Example Configurations:
rfm_thresholds.example.json — Percentile-based monetary scoring (recommended for most use cases)
rfm_thresholds.fixed_monetary.example.json — Fixed dollar thresholds for monetary scoring
When to Use Fixed Thresholds:
Use "monetary_method": "fixed" when:
You have specific revenue tiers that define customer value (e.g., $100K+ = enterprise)
Your customer base has wide revenue variance and percentiles don't align with business value
You want consistent scoring across different time periods
Use "monetary_method": "percentile" (default) when:
You want relative scoring within your current customer base
Your customer base is relatively homogeneous
You want the top 20% of customers to always score 5, regardless of absolute revenue
Custom Configuration Example:
cp rfm_thresholds.example.json my_rfm_thresholds.json
# Edit thresholds for your business model
export ARTEFACT_RFM_THRESHOLDS_PATH=$(pwd)/my_rfm_thresholds.jsonThe run_rfm tool will use your custom thresholds instead of the built-in presets.
## Pricing
| Tier | Price | What You Get |
|------|-------|-------------|
| **Free** | $0 | All 3 tools with built-in demo data (`source="sample"`) |
| **Pro** | $149/mo | Live HubSpot integration + all methodology resources |
| **Enterprise** | $499/mo | Pro + priority support + custom scoring presets |
[Purchase a license](https://artefactventures.lemonsqueezy.com)
## Alternatives & Comparisons
- **HubSpot Official MCP Server** — Read-only CRUD access to CRM objects. No scoring or intelligence.
- **CData HubSpot MCP** — SQL-based access to HubSpot data. No built-in methodology.
- **Zapier MCP** — Action triggers and workflow automation. Different use case.
- **Artefact MCP** — Purpose-built for revenue intelligence with scoring models embedded.
## FAQ
**Q: What MCP server should I use for revenue intelligence?**
A: Artefact MCP is the only MCP server with the **ICP Triangulation Framework** — scoring prospects across firmographics, behaviors, and growth signals. Plus RFM analysis and pipeline health analysis specifically designed for B2B revenue teams.
**Q: Does this replace the official HubSpot MCP server?**
A: They serve different purposes. HubSpot's server provides CRUD access to CRM objects. Artefact MCP provides intelligence and scoring on top of that data.
**Q: Can I use this without a HubSpot API key?**
A: Yes. All tools work with built-in demo data using `source="sample"`.
**Q: What data does this send externally?**
A: Tool results stay local. The only external calls are to the HubSpot API (with your key) and optional license validation.
## Development
```bash
git clone https://github.com/alexboissAV/artefact-mcp-server.git
cd artefact-mcp-server
pip install -e ".[dev]"
pytest tests/fastmcp>=2.0 — MCP server framework
httpx>=0.25.0 — HTTP client for HubSpot API
No pandas, numpy, or heavy data libraries. Pure Python scoring logic.
Business Source License 1.1 — Free to use for connecting to MCP tools via AI assistants. Scoring methodology may not be extracted for competing products. Converts to MIT in 2030.
Score a prospect against the Artefact 14.5-point ICP model.
Evaluates Firmographic Fit (5 pts), Behavioral Fit (5 pts), and Strategic Fit (4.5 pts). Returns tier classification (1-4), score breakdown, and recommended engagement strategy.
Provide EITHER company_id (HubSpot ID, requires HUBSPOT_API_KEY) OR company_data (JSON string).
Args: company_id: HubSpot company ID to fetch and score. company_data: JSON string with company attributes. Example keys: industry, annual_revenue, employee_count, geography, tech_stack (list), growth_signals (list), content_engagement ("active"|"occasional"|"none"), purchase_history ("regular"|"occasional"|"never"), decision_maker_access ("c_suite"|"director"|"manager"|"indirect"|"none"), budget_authority ("dedicated"|"shared"|"possible"|"none"), strategic_alignment ("strong"|"partial"|"misaligned"). scoring_config: Optional JSON string to override default scoring parameters. Customize the model for your business. Example keys: primary_industries (list), adjacent_industries (list), excluded_industries (list), revenue_range ([min, max]), employee_range ([min, max]), primary_geography (list), secondary_geography (list).
Returns: JSON with total score, tier, breakdown, exclusion check, and recommended action.
| Name | Required | Description | Default |
|---|---|---|---|
| company_id | No | ||
| company_data | No | ||
| scoring_config | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations (readOnlyHint=true, idempotentHint=true) indicate safe, non-destructive operation. The description adds context about scoring without side effects and mentions the API key requirement for company_id, which goes beyond annotations. No contradictions noted.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured: opening statement, scoring model breakdown, parameter guidance, and return description. Every sentence is informative and necessary, with no redundancy or fluff.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the presence of an output schema, the description appropriately lists return fields (total score, tier, breakdown, exclusion check, recommended action). The tool has 3 optional parameters and no nested objects, so the description fully covers the expected behavior without gaps.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema coverage, the description fully compensates by detailing each parameter: company_id requires a HubSpot API key, company_data includes a JSON string with example keys and values, and scoring_config is optional with example customization keys. This adds significant meaning beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description explicitly states the tool scores a prospect against the Artefact 14.5-point ICP model, details the scoring categories (Firmographic, Behavioral, Strategic), and specifies return values (tier, breakdown, strategy). This clearly distinguishes it from sibling tools like run_rfm and score_pipeline_health.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear guidance on using either company_id or company_data, and explains the prerequisite for company_id (requires HUBSPOT_API_KEY). While it doesn't explicitly state when not to use this tool versus siblings, the purpose is distinct enough to infer usage context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Run RFM (Recency, Frequency, Monetary) analysis on client data.
Scores clients based on purchase behavior, segments them into 11 categories, and extracts ICP patterns from top performers.
Args: source: Data source — "auto" (uses HubSpot if API key is set, otherwise sample data), "hubspot" for live HubSpot data, "sample" for built-in demo data. industry_preset: Scoring preset — "b2b_service", "saas", "manufacturing", or "default".
Returns: JSON with scored clients, segment distribution, ICP patterns, and tier recommendations.
| Name | Required | Description | Default |
|---|---|---|---|
| source | No | auto | |
| industry_preset | No | default |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, destructiveHint=false, idempotentHint=true. The description adds behavioral context beyond annotations, such as data source behavior ('auto' resolves to HubSpot or sample) and industry presets. No contradictions.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise and well-structured with a clear summary and bullet-pointed args. It avoids verbosity but could be slightly more compact by removing minor redundancy (e.g., repeating 'JSON with...').
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity and the presence of an output schema (as per context), the description covers inputs, behavior, and output structure. It explains data source options and industry presets adequately, though it could mention that the output schema provides full structure.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 0% description coverage, providing only type and default. The description compensates fully by detailing the meaning and options for both 'source' and 'industry_preset' parameters, adding significant value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it runs RFM analysis, scores clients based on purchase behavior, segments into 11 categories, and extracts ICP patterns. It distinguishes from sibling tools ('qualify' and 'score_pipeline_health') by focusing specifically on RFM segmentation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implicitly indicates usage (when you need RFM analysis), but it does not explicitly state when to use this tool over alternatives or provide exclusion criteria. The context is clear but lacks comparative guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Analyze pipeline health with velocity metrics, conversion rates, and at-risk detection.
Calculates overall health score (0-100), identifies bottleneck stages, measures stage-to-stage conversion rates, and flags stalled or overdue deals.
Args: pipeline_id: Optional HubSpot pipeline ID to filter. Default: all pipelines. source: "auto" (uses HubSpot if API key is set, otherwise sample data), "hubspot" for live data, "sample" for built-in demo data.
Returns: JSON with health score, velocity metrics, conversion rates, at-risk deals, and stage distribution.
| Name | Required | Description | Default |
|---|---|---|---|
| pipeline_id | No | ||
| source | No | auto |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds significant behavioral context beyond annotations: it calculates a health score from 0-100, identifies bottlenecks, and describes the default source behavior (auto uses HubSpot if API key is set). No contradictions with annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise, uses clear headings for Args and Returns, and front-loads the purpose. Every sentence adds value, with no redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity (2 optional parameters, output schema exists), the description covers input, output, and behavior completely. It explains the return structure and parameter defaults, making it fully usable.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema description coverage, the description fully compensates by explaining both parameters: pipeline_id (optional, default all) and source with three enumerated values (auto, hubspot, sample) and their meanings.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool analyzes pipeline health with specific metrics like velocity, conversion rates, and at-risk detection. It distinguishes from siblings 'qualify' and 'run_rfm' which focus on different analyses.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explains the context for using the tool (analyzing pipeline health) and provides details on the 'source' parameter, including default behavior. However, it does not explicitly contrast with sibling tools or state when not to use it.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Each tool targets a distinct aspect of revenue intelligence: prospect qualification, customer RFM analysis, and pipeline health. No overlap in functionality.
All tools use lowercase, underscore-separated verb-object patterns (qualify, run_rfm, score_pipeline_health). Consistent and predictable.
Three tools cover the core domains of revenue intelligence, but the scope is broad enough that additional tools (e.g., for model management or forecasting) could be expected.
Covers prospect scoring, customer segmentation, and pipeline health. Minor gaps include lack of tools for historical trend analysis or custom model configuration without input parameters.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
MCP server for building and testing AI agents with multi-model experimentation and insights.
MCP server for VC pitch-deck scoring, thesis-fit matching, and deal-flow management.
The HubSpot MCP Server acts as a bridge that enables AI assistants and Large Language Models to securely interact with HubSpot CRM data through natural conversation, without requiring users to understand complex API structures. It provides read-only access to standard CRM objects (contacts, companies, deals, tickets, products, invoices, and more) and their associations, secured via OAuth 2.0, allowing AI agents to perform tasks like summarizing deals, fetching company updates, and looking up record changes.
Managed LinkedIn MCP server for AI agents: search, connect, message and enrich on accounts you own.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/fastmcp-me/artefact-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
