CRM MCP Tools Reference

The Hyperleap CRM MCP provides 9 tools that read your CRM data. You don't need to call tools by name — just ask Claude what you want in natural language and it will select the right tool automatically.

Tools Overview

get_crm_dashboard

Get a single-glance overview of your entire CRM. This is the best starting point for daily check-ins or standup prep.

What it returns

• Total lead count
• New leads added this week
• Lead distribution across pipeline stages (with stage colors)
• Lead distribution across chatbots (which bots are generating leads)
• Recent CRM activity feed (stage changes, insight extractions, etc.)
• Conversion rate, won count, and lost count

Example prompts

• "Show me my CRM dashboard"
• "How many new leads did I get this week?"
• "What's my conversion rate?"
• "Which chatbot is generating the most leads?"

get_pipeline_stages

See how your sales pipeline is structured and where leads are concentrated. Useful for identifying bottlenecks.

What it returns

• All configured stages (default: New → Contacted → Qualified → Proposal → Won → Lost)
• Stage colors and display order
• Which stage is the default for new leads
• Which stages count as Closed-Won vs Closed-Lost
• Lead count in each stage

Example prompts

• "What are my pipeline stages?"
• "How many leads are in each stage?"
• "Where are the bottlenecks in my pipeline?"

list_leads

Browse and search your leads with powerful filtering. Results are paginated with up to 50 leads per page.

What it returns

For each lead: name, email, phone, company, current pipeline stage, tags, source chatbot, last activity date, top insights, and created/modified timestamps.

Filtering options

• Search — by name, email, or phone number
• Chatbot filter — show only leads from a specific chatbot
• Stage filter — show only leads in a specific pipeline stage
• Tag filter — filter by one or more tags (comma-separated)
• Pagination — navigate through results, up to 50 per page

Example prompts

• "Show me all leads from the Support Bot chatbot"
• "Find leads tagged 'enterprise'"
• "Who's stuck in the Contacted stage?"
• "Search for leads with @example.com emails"
• "Show me all qualified leads"

get_lead_details

Get the full 360° view of a single lead. Ideal for pre-call preparation or understanding a lead before reaching out.

What it returns

• Contact info (name, email, phone, country code)
• Company name
• Tags and source information
• Current pipeline stage
• Custom data captured from the chatbot lead form
• All AI-extracted insights (with category, confidence, and source)
• Notes attached to the lead
• Activity history
• Source chatbot
• Created and modified timestamps

Example prompts

• "Tell me everything about jane@example.com"
• "Show me the full profile for this lead"
• "What do we know about Acme Inc?"

get_lead_activities

View the chronological activity log for a specific lead. Useful for understanding what's happened recently or preparing for a handoff between team members.

What it returns

• Stage changes (with from → to metadata)
• Notes added
• Insights extracted
• Conversation events
• Who performed each action and when

Example prompts

• "What's happened with this lead recently?"
• "Show me the activity timeline for this lead"
• "When was this lead last updated?"

get_lead_notes

Pull all human-written notes attached to a lead. These are notes your team has manually added, not AI-generated content.

What it returns

• Note content
• Author who wrote the note
• Created and modified timestamps

Example prompts

• "Show me all notes for this lead"
• "What has the team written about this lead?"
• "Read me the sales notes for jane@example.com"

get_lead_conversations

List the conversations a lead has had with your chatbot. It uses the chatbot the lead came in through, unless you name a different one. Your assistant usually follows up with get_conversation to read the ones that matter.

What it returns

• Each conversation's ID and name
• Message count
• Conversation status

Example prompts

• "Which conversations has this lead had?"
• "List jane@example.com's chats with the Support Bot"

get_conversation

Read one conversation in full. See exactly what the lead asked about and how the AI responded.

What it returns

• Conversation name, status, and message count
• AI model used
• Every visitor and AI message, with timestamps

Example prompts

• "What did this lead ask about?"
• "Replay the chat for jane@example.com from the Support Bot"

extract_lead_insights

Trigger AI analysis across a lead's conversations to surface structured insights. This actively processes the lead's data rather than just retrieving it.

What it returns

• Contact information (name, email, phone, country)
• Location data
• Topics discussed
• User intent and sentiment
• Each insight tagged with category and confidence score

Example prompts

• "Extract insights from this lead's conversations"
• "Analyze this lead's intent and sentiment"
• "What topics did this lead discuss?"

Common Workflows

Daily CRM Review

Start with the dashboard for a high-level overview, then drill into new leads or pipeline bottlenecks:

1. "Show me my CRM dashboard"
2. "List the new leads from this week"
3. "Tell me more about [specific lead]"

Pre-Call Research

Prepare for a sales call by getting the full picture of a lead:

1. "Show me everything about jane@example.com"
2. "What notes has the team written about this lead?"
3. "Show me their conversation history"
4. "Extract insights from their conversations"

Pipeline Analysis

Understand your pipeline health and identify stuck leads:

1. "What are my pipeline stages and how many leads are in each?"
2. "Show me all leads stuck in the Contacted stage"
3. "What's my conversion rate?"

Limitations

• Reads, with one exception — You cannot create leads, update pipeline stages, add notes, delete records, or send messages through the MCP; do those in Hyperleap Studio. The only thing the MCP saves is extract_lead_insights adding insights to the lead's profile, plus an activity-log entry that it ran.
• Your access, not more — An MCP key acts as you: it sees the leads and chatbots you can see in Studio. Anyone holding the key sees the same, so don't share it.
• Real-time data — Results reflect the current state of your CRM. There is no historical snapshot or time-travel capability.

Next Steps

• Connect CRM to AI Assistants — Set up the MCP connection if you haven't already
• Webhooks — Set up real-time notifications when chatbot events occur
• API Reference — Full REST API documentation for programmatic access
• The 9 Hyperleap MCP tools, explained with real prompts — Pillar reference: parameters, example prompts, sample responses, when to use each