Official Grain MCP Server: Quick Start Guide

The Grain MCP (Model Context Protocol) connects Claude.ai (and other AI chat clients) directly to your meeting data in Grain, enabling powerful AI-powered individual workflows as well as customer and team analytics/reporting directly within Claude.ai. While we are working on support for ChatGPT (email [mike (at) grain (dot) com] if you want to be on the ChatGPT early access list), we encourage users who plan to prioritize MCP in their workflow to use Claude as it's support for MCP is currently meaningfully more mature and robust.

Claude 60-Second Setup

Getting Started Video:

In this video I walk through the set-up steps that are also documented below.

Please note that if you're using the Teams or Enterprise versions there is a slight modification of the "Add Integration" step detailed in the first section below.

Claude Paid Users

Grain MCP supports easy setup as a native Claude Integration.

Step 1: Integration Setup

1. Navigate to Settings → Integrations (Pro/Max Users) → Organization Settings (Team/Enterprise Users)

1. Click "+ Add Integration"

1. Enter details and click the "Add" button:

• Integration Name: Grain
• Integration URL: https://api.grain.com/_/mcp

Step 2: Connection / Authentication

From Settings → Integrations:

1. Look for "Grain" in the list of integrations and click "Connect"

1. Once automatically redirected to Grain, click "Approve"

Step 3: Testing MCP Tools

From a New Claude chat:

1. Click "Search and Tools" and confirm you can see "Grain" in the list

1. Click "Grain" to confirm the MCP tools are active, toggle tools on/off as desired

1. Test Connection by typing "Tell me about my most recent Grain meeting" or any other test prompt and Submit the chat

If everything is set up properly, you'll see Claude make requests to the available Grain tools that are needed to fulfill the data needs of the Chat request. For best results, turn off non-Grain tools that may distract or confuse the LLM or cite the name and filters of the Grain tool you want it to use from the Available MCP Tools section.

Claude Free Users & Other Clients (Developers Only)

Cursor

Click Here to Install

Manual Configuration (All Other Clients)

Prerequisites:

• Node.js installed (nodejs.org)
• Claude Desktop App

Verify Node.js Installation:

Open terminal/command prompt and run:

node --version

If you get an error, install Node.js from nodejs.org.

Setup Steps:

1. Edit Config File

   • Settings → Developer → Edit Config
   • Opens claude_desktop_config.json

2. Add Configuration

   {
     "mcpServers": {
       "grain": {
         "command": "npx",
         "args": ["-y", "mcp-remote", "https://api.grain.com/_/mcp"]
       }
     }
   }

3. Restart Application

   • Close and reopen Claude Desktop
   • Verify "Grain" appears in Settings → Integrations

4. Enable Tools

   • Select Tools and Settings for Grain
   • Verify all tools are enabled
   • Allow tools when prompted in conversations

Test Connection:

List my recent Grain meetings

ChatGPT Quick Setup

1. Go to **chat.openai.com** and sign into your ChatGPT Pro account

1. Go to settings and navigate to "Connectors"

1. Set up Grain as a connector using the url https://api.grain.com/_/mcp

1. Grant ChatGPT access to your Grain Account data

1. Enable Grain as a source and begin your work!

Note: you will need to open a new chat and enable Grain as a connector before beginning each query. If you have already started a chat, Grain will be greyed out as a connector option.

Available MCP Tools

Free & Starter Plan MCP Tools

• myself : Get your Grain account information

  Parameters: None

• list_meetings / list_attended_meetings: Get filtered list of all accessible meetings

  Key Parameters:

  • filters (optional): Object containing:
    • companies: Array of company IDs (use search_companies first)
    • persons: Array of person IDs (use search_persons first)
    • after_datetime: ISO 8601 formatted datetime (e.g., "2024-01-01T00:00:00Z")
    • before_datetime: ISO 8601 formatted datetime
    • participant_scope: "internal" | "external"
    • title_search: Substring to match in meeting titles
  • limit: Results per page (1-20, default 10)
  • cursor: For pagination

• fetch_meeting: Get detailed information about a specific meeting

  Key Parameters:

  • meeting_id: UUID of the meeting

• fetch_meeting_transcript: Retrieve full meeting transcript

  Key Parameters:

  • meeting_id: UUID of the meeting
  • include_timestamps: Boolean (default: false)

• fetch_meeting_notes: Get AI-generated meeting notes (more concise than transcripts)

  Parameters:

  • meeting_id: UUID of the meeting

• search_meetings: Semantic search across all meeting transcripts

  Key Parameters:

  • search_string: Your search query (longer strings = more focused results)
  • filters: Same meeting filters as list_meetings
  • limit: Results to return (1-50, default 10)
  • speaker_scope: "all" | "internal" | "external" (default: all)

• search_companies: Find companies that participated in meetings when filtering

  Key Parameters:

  • search_string: Company name or domain
  • filters: Standard meeting filters
  • limit: Results to return (1-20, default 10)

• search_persons : Search for meeting participants when filtering

  • search_string: Person's name or email
  • filters: Standard meeting filters
  • limit: Results to return (1-20, default 10)

• list_workspace_users : Get all users in your Grain workspace

  Parameters: None

Business & Enterprise Plan MCP Tools

• list_coaching_feedback : Get AI-generated sales coaching insights

  Key Parameters:

  • filters.has_coaching_opportunities: Set to true for flagged opportunities only
  • Standard meeting filters apply

• fetch_meeting_coaching_feedback : Get detailed coaching scorecard for specific meeting

  Parameters:

  • meeting_id: UUID of the meeting

• list_open_deals / list_all_deals : Access HubSpot Deal intelligence

  Key Parameters:

  • filters (optional): Object containing:
    • companies: Array of company IDs (UUIDs)
    • deal_at_risk: Boolean to filter at-risk deals
    • deal_owner: Person ID (UUID) of deal owner
    • pipeline: HubSpot pipeline ID
    • title_search: Substring to match in deal titles
  • limit (optional): Results per page (1-20, default 10)
  • cursor (optional): For pagination

• fetch_deal : Get detailed deal information

  Parameters:

  • deal_id: UUID of the deal

Basic Usage Examples

Start with these foundational prompts to explore your meeting data.

Getting Started (First Steps)

• Show me all action items I'm responsible for from the past 7 days
• Help me draft all follow-up emails for my external meetings today

Customer Intelligence

• Summarize customer mentions of pricing concerns in the last 30 days
• Help me understand how customers are comparing us with competitors

People & Company Insights

• Who from {Company Name} have we met with and what topics did we cover?
• Show me all users who mentioned feature requests or product feedback recently

Deal & Business Analysis

• What objections came up in external sales calls this week?
• Show me deals that haven't had meetings in the last 2 weeks

Content Discovery

• Search for discussions about our new product features
• What were the key decisions made in this week's internal meetings?

Structured Reporting Examples

Using MCP Prompts

While MCP Tools are designed to work with any natural language inputs into Chat, Prompts are manually selected/triggered by the user to chain multiple tool calls together in a structured and repeatable way. The result is a set of comprehensive analysis workflows that would take hours manually but complete in minutes. We have created a initial library of 9 officially supported Prompts accessible directly within Claude.ai, you can riff on our templates to create your own prompts by copy/pasting your own changes to the prompt's instructions.

How to Use:

1. Click the + Button in the bottom left of the Claude chat window
2. Click "Add from Grain" in the options list and choose a report to run
3. Submit Prompt to Claude with any (optional) custom instructions in the chat window
4. Input Variables when prompted by Claude as needed

Product & Market Intelligence

Broad market insights and specific feature/competitive deep-dives

Voice of Customer Insights Report

Prompt Name: voice_of_customer_insights_report

Use for: Product roadmap priorities, market trends, customer satisfaction

Arguments:

• analysis_days (optional): Days of customer feedback to analyze (default: 120)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. list_meetings - Get external meetings with customers
2. search_meetings - Find pain points, feature requests, and success stories
3. search_companies - Segment feedback by customer type
4. fetch_meeting_notes - Extract detailed insights and quotes

Copy/Paste the below into any AI Chat Client w/ Grain MCP active or use Prompts in Claude.ai

Extract customer insights across all feedback channels over the last {analysis_days*} days:

Step 1: Comprehensive data collection
- Use list_meetings* with participant_scope=external to find external meetings with customer pain points mentioned
- Use search_meetings *to find feature requests and enhancement discussions
- Use search_meetings* to identify implementation challenges and successes
- Use search_meetings *to locate competitive comparisons and concerns

Step 2: Pain point categorization:
- Use search_meetings* to categorize product functionality gaps
- Use search_meetings *to identify integration limitations
- Use search_meetings* to find user experience friction
- Use search_meetings *to document performance or reliability issues
- Use search_meetings* to extract pricing and value concerns

Step 3: Success story identification:
- Use search_meetings *to find positive outcome testimonials
- Use fetch_meeting_notes to extract ROI and value realization examples
- Use search_meetings* to document workflow improvements achieved
- Use search_meetings *to identify team adoption successes

Step 4: Feature demand analysis:
- Analyze request frequency by feature category
- Use search_companies* to identify customer segment patterns
- Assess business impact and urgency levels
- Evaluate technical complexity vs. customer value

Step 5: Actionable insights:
- Generate product roadmap priorities with customer backing
- Identify customer success best practices
- Recommend support process improvements
- Highlight competitive positioning opportunities

Include specific customer quotes and quantified impact data.

Example of reporting output from the prompt (redacted and shortened)

Deal Intelligence

Big picture view of all deals and individual deal deep-dives.

[All reports require Grain Business or Enterprise account with an active HubSpot Integration]

Deal Pipeline Intelligence Report

Prompt Name: deal_pipeline_intelligence_report

Use for: Portfolio overview, resource allocation, trend analysis

Arguments:

• days_back (required): Number of days to analyze (default: 60)
• risk_threshold (optional): Deal age threshold for high risk classification (default: 15)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. list_open_deals - Get all active deals
2. list_meetings - Find meetings for each deal
3. fetch_meeting_notes - Analyze discussion content
4. search_meetings - Find competitive mentions and objections

*Analyze active deal health and progression over the last {days_back} days:

Step 1: Get all open deals with recent activity
- Use list_open_deals to retrieve active opportunities
- Include deal metadata: stage, momentum, risk status, scores

Step 2: For each deal with external meetings:
- Use fetch_deal filtered by company to get meeting history
- Analyze deal age and progression velocity
- Review stakeholder engagement patterns
- Assess meeting frequency trends vs. stage requirements
- Use fetch_meeting_notes to extract key discussion topics and pain points
- Use search_meetings for competitive mentions and objections
- Evaluate next steps clarity and follow-through

Step 3: Risk assessment by category:
- HIGH RISK: Stalled momentum, {riskthreshold}+ days old, low scores
- MEDIUM RISK: Slowing momentum, integration concerns
- HEALTHY: Progressing momentum, recent activity, clear next steps

Step 4: Generate actionable recommendations:
- Immediate interventions for at-risk deals
- Expansion opportunities in healthy accounts
- Specific next steps with timelines
- Resource allocation priorities

Include deal IDs, company names, and specific quotes from meetings.*

Health Report By Deal

Prompt Name: health_report_by_deal

Use for: Specific deal strategy, stakeholder mapping, next step planning

Arguments:

• deal_name (required): Deal name or company name to analyze
• analysis_days (optional): Days of meeting history to analyze (default: 90)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. search_companies - Find company ID by name
2. fetch_deal - Get deal metadata if available
3. list_meetings - Get all meetings with the company
4. fetch_meeting_notes - Extract key insights from each meeting
5. search_meetings - Find specific topics like competitors or objections

Analyze the health and momentum of {deal*name}:

Step 1: Deal identification and context
- Use search_companies* to find company ID for "{deal*name}"
- Use fetch*deal to retrieve deal metadata: stage, age, owner, risk status, scores
- Use search*persons to identify key stakeholders and contacts involved

Step 2: Meeting Analysis Sources table
Create a reverse chronological table with:
- Date, Meeting Title, Duration
- Attendees (exclude bot/notetaker names)
- Upshot: 15-word-or-less summary of purpose and outcome
- Source: Percentage indicating relative importance to analysis

Step 3: Meeting activity analysis
- Use fetch_deal* filtered by company to find all associated meetings
- Analyze meeting frequency and timing patterns over last {analysis_days*} days
- Review participant engagement and attendance
- Track progression through sales stages

Step 4: Content analysis of conversations
- Use fetch_meeting_notes to extract key discussion topics and pain points
- Use search_meetings* to identify decision criteria and evaluation process
- Use search_meetings for competitive mentions and objections
- Note buying signals and momentum indicators

Step 5: Stakeholder mapping and engagement
- Map decision makers, influencers, and champions from meeting attendees
- Analyze engagement levels of key participants
- Identify missing stakeholders or access gaps
- Review multi-threading effectiveness

Step 6: Risk assessment and recommendations
- Evaluate deal progression velocity vs. stage requirements
- Identify stall indicators or momentum shifts
- Assess competitive threats and positioning
- Generate specific next steps with timelines

Include deal score trends, specific quotes, and actionable recommendations.

SPICED Opportunity Matrix By Deal

Prompt Name: spiced_opportunity_matrix_by_deal

Use for: Comprehensive deal qualification, stakeholder alignment, forecasting accuracy

Arguments:

• deal_name (required): Deal name or company name to analyze
• analysis_period (optional): Days of meeting history to analyze (default: 180)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. search_companies - Find company by name
2. fetch_deal - Get deal metadata and scores
3. list_meetings - Get comprehensive meeting history
4. fetch_meeting_transcript - Deep content analysis for SPICED elements
5. search_meetings - Find specific SPICED criteria discussions

Generate a complete SPICED analysis for {deal*name}:

Step 1: Deal identification and meeting collection
- Use search_companies* to find company ID for "{deal*name}"
- Use list_open_deals and fetch_deal to get deal metadata if available
- Use fetch*deal to get deal metadata if available
- Use list_meetings *to find all meetings with this company over {analysis_period*} days
- List participants and their roles across all meetings
- Create timeline of interactions and progression

Step 2: Meeting Analysis Sources table
Create a reverse chronological table with:
- Date, Meeting Title, Duration
- Attendees (exclude bot/notetaker names)
- Upshot: 15-word-or-less summary of purpose and outcome
- Source: Percentage indicating relative importance to SPICED analysis

Step 3: SITUATION analysis from meeting content
Use fetch*meeting*transcript to extract and document background facts:
- Company size, locations, and organizational structure mentioned
- Current technology stack and existing solutions
- Recent initiatives, changes, or strategic shifts discussed
- Industry context and market position
- Key business metrics or KPIs referenced

Step 4: PAIN identification and prioritization
Use search_meetings *to analyze all meetings for pain points:
- Quantifiable pains: Measurable business challenges with metrics
- Qualitative pains: Emotional or subjective challenges expressed
- Pain evolution: How pains have changed/evolved across meetings
- Pain prioritization: Which pains are mentioned most frequently
- Direct quotes expressing frustration or challenges

Step 5: IMPACT assessment and business case
Use fetch*meeting_notes to document desired outcomes and value:
- Revenue increase opportunities mentioned
- Cost reduction possibilities discussed
- Customer experience improvements sought
- Individual/emotional impacts for stakeholders
- ROI discussions and value quantification attempts
- Success metrics and KPIs they want to achieve

Step 6: CRITICAL EVENT timeline and urgency
Identify all timing drivers:
- Hard deadlines mentioned (board meetings, renewals, launches)
- Compelling events creating urgency (hiring, compliance, competition)
- Consequences of missing deadlines explicitly discussed
- Timeline evolution: How dates have shifted across meetings
- Test questions asked about timing importance

Step 7: DECISION process mapping
Comprehensive decision analysis:
- Decision process: Steps and stages they've outlined
- Decision committee: All stakeholders mentioned and their roles
  - Economic buyer identification
  - Technical evaluators
  - Executive sponsors
  - Influencers and blockers
- Decision criteria: Evaluation factors discussed
  - Technical requirements
  - Integration needs
  - Budget constraints
  - Security/compliance requirements
- Previous purchase experiences mentioned
- Procurement/legal requirements identified

Step 8: Competitive landscape and status quo
- Current solution pain points and limitations
- Competitors being evaluated (direct mentions)
- Evaluation criteria comparisons
- Differentiation points discussed
- Risk of no decision vs. competitive loss

Step 9: SPICED opportunity matrix generation
Create a structured summary table:
- Situation: 3-5 key facts
- Pain: Top 3 quantifiable and qualitative pains
- Impact: Primary business impacts with metrics
- Critical Event: Specific date and consequences
- Decision: Process flowchart and committee map

Step 10: Deal qualification and forecast accuracy
- SPICED completeness score (what's known vs. unknown)
- Strength of pain/impact connection
- Critical event credibility assessment
- Decision process clarity and access
- Overall deal qualification rating
- Recommended discovery gaps to fill

Step 11: Strategic recommendations
- Next discovery questions to ask
- Stakeholders to engage based on SPICED gaps
- Value proposition alignment with impacts
- Timeline acceleration opportunities
- Risk mitigation strategies

Include specific quotes for each SPICED element, meeting references, and a visual opportunity matrix summary.

MEDDICC Diagnosis Report By Deal

Prompt Name: meddicc_diagnosis_report_by_deal

Use for: Enterprise deal qualification, complex sales cycles, multi-stakeholder opportunities

Arguments:

• deal_name (required): Deal name or company name to analyze
• analysis_period (optional): Days of meeting history to analyze (default: 180)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. search_companies - Find company by name
2. fetch_deal - Get deal metadata and qualification scores
3. list_meetings - Get comprehensive meeting history
4. fetch_meeting_transcript - Deep content analysis for MEDDICC elements
5. search_meetings - Find specific qualification criteria
6. search_persons - Map stakeholders and decision makers

Generate a complete MEDDICC analysis for {deal*name}:

Step 1: Deal identification and meeting collection
- Use search_companies* to find company ID for "{deal*name}"
- Use list_open_deals and fetch_deal to get deal metadata if available*
- Use list_meetings *to find all meetings with this company over {analysis_period*} days
- Use search_persons to list participants and their roles across all meetings
- Create timeline of interactions and progression

Step 2: Meeting Analysis Sources table
Create a reverse chronological table with:
- Date, Meeting Title, Duration
- Attendees (exclude bot/notetaker names)
- Upshot: 15-word-or-less summary of purpose and outcome
- Source: Percentage indicating relative importance to MEDDICC analysis

Step 3: METRICS identification and quantification
Extract measurable success criteria:
- Specific KPIs and targets mentioned
- Current performance baselines discussed
- Improvement goals with numbers and timelines
- ROI expectations and calculations shared
- Success measurement criteria defined

Step 4: ECONOMIC BUYER mapping
Identify and analyze the economic buyer:
- Name, title, and direct influence confirmed
- Budget authority validation
- Meeting participation and engagement level
- Direct quotes showing financial decision power
- Access level achieved and relationship strength

Step 5: DECISION CRITERIA documentation
Catalog all evaluation criteria mentioned:
- Technical requirements and specifications
- Business/operational requirements
- Financial constraints and budget parameters
- Timeline and implementation needs
- Competitive evaluation criteria
- Weighting or priority of criteria

Step 6: DECISION PROCESS analysis
Map the complete buying process:
- Formal steps outlined by customer
- Informal political dynamics observed
- Approval stages and sign-offs required
- Committee members and voting process
- Timeline for each stage
- Current stage and next steps

Step 7: IDENTIFY PAIN deep dive
Comprehensive pain analysis:
- Business pains with quantified impact
- Technical pains affecting operations
- Personal pains for key stakeholders
- Pain severity and urgency ratings
- Consequences of inaction discussed
- Pain ownership and accountability

Step 8: CHAMPION qualification
Assess champion strength and capability:
- Name and role of potential champions
- Evidence of internal selling activities
- Power and influence demonstrated
- Personal win if project succeeds
- Ability and willingness to provide intel
- Risk if champion is removed/changes

Step 9: COMPETITION assessment
Analyze competitive landscape:
- Direct competitors being evaluated
- Incumbent solution strengths/weaknesses
- Competitive advantages articulated
- Vulnerabilities and differentiation points
- Decision criteria favoring us vs. them
- Competitive strategy and positioning

Step 10: MEDDICC scoring and gaps
Create qualification scorecard:
- Score each MEDDICC element (1-5)
- Identify missing information per element
- Risk assessment for weak areas
- Strength indicators for leverage
- Overall deal qualification score

Step 11: Strategic action plan
Based on MEDDICC analysis:
- Critical gaps requiring immediate discovery
- Stakeholder engagement priorities
- Champion development strategies
- Competitive positioning tactics
- Risk mitigation approaches
- Specific next steps with owners and dates

Include quantified metrics, specific quotes, and a MEDDICC scorecard summary.

Account & Customer Health

Cross-account trends and individual account deep-dives

Recently Active Accounts Health Report

Prompt Name: recently_active_accounts_health_report

Use for: Portfolio triage, expansion opportunities, churn risk identification

Arguments:

• analysis_days (optional): Days to analyze meeting activity (default: 90)
• account_count (optional): Number of top accounts to analyze (default: 5)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. list_meetings - Get meetings with external participants
2. search_companies - Identify most active companies
3. fetch_meeting_notes - Analyze sentiment and topics
4. search_meetings - Find risk indicators and concerns

Analyze the health of my top {*account*_*count} accounts based on recent meeting activity:

1. Use list_meetings* with external participant filter to identify companies with the most meetings in the last {analysis_days*} days
2. For each company, use multiple tools to analyze:
   - Use list_meetings* to assess meeting frequency trends
   - Use *search*_*persons to review participant engagement levels
   - Use fetch_meeting_notes to extract key topics and concerns discussed
   - Use search_meetings* to track sentiment progression over time
   - Use search_meetings to identify risk indicators (missed meetings, unresolved issues)
3. Provide specific recommendations for each account
4. Flag any accounts that need immediate attention

Focus on companies where we've had external participants.

Customer Health Report By Account

Prompt Name: customer_health_report_by_account

Use for: Account strategy, renewal planning, expansion tactics

Arguments:

• account_name (required): Company name to analyze
• analysis_period (optional): Days of history to analyze (default: 90)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. search_companies - Find company by name and get contacts
2. list_meetings - Get comprehensive meeting history
3. search_persons - Map stakeholder relationships
4. fetch_meeting_notes - Analyze sentiment and topics
5. search_meetings - Track specific themes and concerns

Comprehensive health analysis for {account_*name}:

Step 1: Account identification and baseline
- Use search_companies* to find company ID for "{account_*name}" and associated contacts
- Use list_meetings* to establish relationship timeline and tenure
- Identify account type (customer, prospect, partner) from meeting patterns
- Review account expansion history and potential from meeting progression

Step 2: Meeting Analysis Sources table
Create a reverse chronological table with:
- Date, Meeting Title, Duration
- Attendees (exclude bot/notetaker names)
- Upshot: 15-word-or-less summary of purpose and outcome
- Source: Percentage indicating relative importance to analysis

Step 3: Meeting engagement analysis ({analysis_period*}+ day period)
- Use list_meetings* to assess meeting frequency trends and consistency
- Use search_*persons to analyze stakeholder participation patterns
- Use fetch_*meeting_notes to review topic evolution and engagement depth
- Compare internal vs external meeting ratios

Step 4: Sentiment and satisfaction tracking
- Progression of sentiment over time
- Support issues and resolution patterns
- Feature requests and enhancement discussions
- Success stories and positive outcomes

Step 5: Business health indicators
- Usage growth or decline signals
- Team expansion or contraction mentions
- Budget and procurement discussions
- Renewal timeline and risk factors

Step 6: Relationship strength assessment
- Champion identification and engagement
- Executive access and relationship depth
- Cross-departmental usage and adoption
- Referral potential and advocacy signals

Step 7: Action plan and risk mitigation
- Immediate retention risks requiring intervention
- Expansion opportunities with specific tactics
- Relationship strengthening recommendations
- Timeline for next touchpoints and objectives

Focus on behavioral indicators, usage signals, and relationship quality metrics.

SPICED Opportunity Matrix by Customer Account

Prompt Name: spiced_opportunity_matrix_by_customer_account

Use for: Customer retention strategy, expansion planning, renewal preparation, QBR insights

Arguments:

• account_name (required): Company name to analyze
• analysis_period (optional): Days of history to analyze (default: 180)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. search_companies - Find company by name and get contacts
2. list_meetings - Get comprehensive meeting history
3. search_persons - Map stakeholder relationships
4. fetch_meeting_notes - Analyze sentiment and topics
5. search_meetings - Track specific themes and concerns

Generate a complete SPICED analysis for customer account {account_name}:

Step 1: Account identification and meeting collection
- Search for company name to get company ID and all associated contacts
- Find all meetings with this customer (last {analysis_period} days)
- List participants, roles, and engagement frequency
- Create timeline of touchpoints and support interactions

Step 2: Meeting Analysis Sources table
Create a reverse chronological table with:
- Date, Meeting Title, Duration
- Attendees (exclude bot/notetaker names)
- Upshot: 15-word-or-less summary of purpose and outcome
- Source: Percentage indicating relative importance to customer SPICED analysis

Step 3: SITUATION analysis - Current state assessment
Document the customer's current state:
- Company size, growth, or contraction signals
- Organizational changes (mergers, leadership, restructuring)
- Current product usage patterns and adoption levels
- Integration touchpoints and workflow dependencies
- Team members actively using the solution
- Recent business initiatives or strategic shifts mentioned

Step 4: PAIN identification - Ongoing challenges
Analyze current and emerging customer pains:
- Unresolved technical issues or limitations
- New business challenges since implementation
- Process inefficiencies still present
- Feature gaps causing workarounds
- Support ticket patterns and escalations
- Competitive solutions being evaluated
- Change management struggles

Step 5: IMPACT assessment - Value realization
Measure achieved and potential impacts:
- Quantified value delivered to date
- ROI metrics shared by customer
- Success stories and wins attributed to solution
- Efficiency gains or cost savings realized
- Missed impact opportunities
- Expansion potential if pains addressed
- Risk of churn impact on their business

Step 6: CRITICAL EVENT mapping - Key dates ahead
Identify upcoming critical events:
- Contract renewal date and process
- Budget planning cycles
- Leadership or organizational changes
- Competitive RFP timelines mentioned
- Business seasonality affecting usage
- Product roadmap items they're waiting for
- Expansion project timelines

Step 7: DECISION dynamics - Renewal & expansion
Analyze decision-making for retention/growth:
- Current decision makers vs. original buyers
- New stakeholders to engage
- Budget authority changes
- Procurement process for renewals
- Expansion approval requirements
- Risk factors for non-renewal
- Champion strength and advocacy

Step 8: Customer health scoring
Create SPICED-based health assessment:
- Situation stability (organizational health)
- Pain severity (unresolved issues impact)
- Impact achievement (value realization score)
- Critical event preparedness
- Decision maker engagement level
- Overall retention risk rating

Step 9: Expansion opportunity analysis
Based on SPICED insights:
- Additional use cases to explore
- New teams/departments to engage
- Upsell opportunities aligned to pains
- Cross-sell potential based on impacts
- Reference and advocacy potential
- Case study opportunities

Step 10: Strategic account plan
Develop action plan from SPICED analysis:
- Immediate pain resolution priorities
- Executive business review preparation
- Renewal negotiation strategy
- Expansion conversation timing
- Relationship strengthening tactics
- Competitive defense strategies
- Success metric documentation needs

Include specific quotes showing satisfaction/concern levels, usage trends, and a visual SPICED health matrix.

Sales Performance & Coaching

Team-wide performance patterns and individual rep development

Sales Team Performance & Coaching Opportunities Report

Prompt Name: sales_team_performance_coaching_opportunities_report

Use for: Team training priorities, performance trends, systematic issues

Arguments:

• analysis_days (optional): Days of coaching data to analyze (default: 30)

Implementation:

This prompt uses the following Grain MCP tools in sequence:

1. list_coaching_feedback - Get meetings with coaching opportunities
2. fetch_meeting_coaching_feedback - Get detailed scores for each meeting
3. list_workspace_users - Map reps to their performance data
4. search_meetings - Find patterns in objections and successful calls

Analyze sales performance and coaching needs over the last {analysis_days*} days:

Step 1: Coaching data collection
- Use list_*coaching_*feedback with has_*coaching_*opportunities filter to find flagged meetings
- Use fetch_*meeting_*coaching_*feedback to include overall scores and category breakdowns
- Focus on meetings with external participants (sales calls)

Step 2: Individual performance analysis:
- Use list_*workspace_*users to identify all reps
- Use fetch_*meeting_*coaching_*feedback to track score trends by rep over time
- Analyze weakest skill categories per person
- Review behavioral patterns (talk time, filler words)
- Identify consistent feedback themes

Step 3: Team-wide coaching priorities:
- Aggregate coaching feedback to find most common skill gaps across reps
- Identify systematic issues affecting multiple people
- Use search_meetings* to find best practice examples from high-scoring calls
- Extract role-play scenarios from real objections

Step 4: Development recommendations:
- Generate individual coaching plans with specific focus areas
- Prioritize team training needs with measurable outcomes
- Define success metrics and tracking methods
- Create timeline for improvement assessment

Include specific coaching feedback quotes and score progressions.

Frequently Asked Questions

Q: What applications and platforms support MCP integrations?

A: The Model Context Protocol is rapidly gaining adoption where each client supports different MCP features (tools, prompts, resources) depending on their use cases. For the semi-complete list of Clients with feature compatibility details, visit the official MCP clients directory.

Q: What Claude plan do I need to use MCP integrations?

A: MCP integrations, including the Grain MCP server, are only available on paid Claude plans (Pro, Team, and Enterprise). Free Claude accounts do not support MCP integrations or the ability to connect external tools and data sources. You'll need to upgrade to a paid plan to access the Grain integration and other MCP servers.

Q: Are all Grain MCP tools available on every Grain plan?

A: Most Grain MCP tools work on all Grain plans (Free, Starter, Business, Enterprise), including meeting search, transcripts, notes, and contact management. However, tools for Deals (list_open_deals, list_all_deals, fetch_deal) and Sales Coaching/Scorecards (list_coaching_feedback, fetch_meeting_coaching_feedback) are restricted to Business and Enterprise plan users only. Free and Starter plan users will receive an error if they attempt to use these premium tools.

Q: Is it safe to use MCP servers like Grain's?

A: MCP security relies on a trust model, and Grain's MCP server is officially supported by us and aligns with our existing security practices, follows the MCP specification exactly, and uses industry-standard OAuth authentication and HTTPS. All tool usage requires explicit user approval in Claude, and each tool request is scoped to specific data needed for that operation. While no system is 100% secure, Grain's established security practices and transparent approach to data handling make it a safer MCP implementation.

Q: How does the Model Context Protocol (MCP) work and what are the privacy implications?

A: The Model Context Protocol (MCP) is an open standard that creates a secure bridge between Claude and external services like Grain, where your data stays in your Grain account and is only processed temporarily by Claude to answer questions. As stated by Anthropic, data accessed through MCP integrations is not used to train Claude's models.

Q: What data does Grain's MCP server have access to?

A: Grain's MCP server only accesses data you already have permission to view in your Grain workspace, including meetings you attended, companies and contacts from those meetings, deal information (with HubSpot integration), and AI-generated notes and transcripts. It has no access to private meetings you weren't invited to or other workspaces you don't belong to. The integration uses OAuth authentication and maintains the same security boundaries as your normal Grain usage.

Q: How does Claude verify and approve MCP tool usage?

A: Claude requires explicit user approval before using any Grain tool, showing you a dialog with clear descriptions of what each tool will do and offering "use once" or "use for entire session" options. Each tool requires separate approval when first used, and you maintain control over which tools Claude can access in your current session. You can revoke access by starting a new chat session, ensuring that no tools run without your explicit permission.

Q: Can I control what data the Grain MCP server shares with Claude?

A: Data access is controlled through your existing Grain account permissions, and you can enable/disable specific tools in Claude's integration settings, but you cannot currently exclude specific meetings or companies from MCP access. The integration accesses all data you have permission to see in your Grain workspace and follows your existing permissions rather than providing additional filtering options. Review your Grain workspace permissions before connecting and use targeted prompts to limit the scope of data Claude requests.

Q: What is the helpful_usage_information parameter I see in the client's tool responses?

A: You may notice the helpful_usage_information parameter appearing in chat interface when the Grain MCP server responds to tool requests. This optional parameter helps Grain's team understand tool usage patterns to improve functionality, and this data is never used as AI training data. Currently there's no opt-out option, but one will be available in the coming months - if you're uncomfortable with this data collection, consider waiting until then.