Close MCP
Close MCP gives agents scoped access to CRM leads, contacts, opportunities, activities, communications, and organization workflows.
Verdict
Common use cases
- Pull pipeline snapshots before team standups
- Log call notes and meeting outcomes inline
- Query deal stage and close probability
- Update contact fields from chat context
- Generate weekly forecast summaries for leadership
Integration
- Vendor
- Close MCP
- Category
- other
- Auth
- OAUTH2
- Composio slug
close_mcp
Tools
Setup
Setup guide
- 1In Switchy, open your workspace settings and navigate to Integrations. Click 'Add MCP Server' and select Close from the directory. You'll be redirected to Close's OAuth consent screen — sign in with an account that has API access enabled. Grant the requested scopes: read and write permissions for leads, opportunities, contacts, and activities. After authorizing, Switchy will confirm the connection and display available tools. Return to any Project and type '@Close' followed by a command to test — try asking for a list of open opportunities or recent activity on a specific lead. If the MCP responds with data, the connection is live.
What teammates see: by default, memories from Close MCP are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Pipeline Snapshot
@Close show me all open opportunities grouped by stage, with total value per stage
Recent Lead Activity
@Close get the last five activities for lead ID 12345, including emails and calls
Log Call Notes
@Close log a call activity for contact Jane Doe at Acme Corp: discussed Q2 renewal, next step is pricing proposal by Friday
Update Contact Field
@Close update contact john@startup.io to set the 'Industry' field to SaaS and 'Employee Count' to 50
Forecast Summary
@Close calculate weighted pipeline value for opportunities closing this quarter, grouped by month
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Close pull all closed-won deals from Q4 2024 and summarize the average deal size, most common lead source, and longest sales cycle
Retrieved 34 closed-won deals from Q4 2024. Average deal size: $47,200. Most common lead source: referral (15 deals), followed by inbound marketing (11 deals). Longest sales cycle: 127 days for the deal with MegaCorp Industries, which moved through 8 pipeline stages. Shortest cycle: 12 days for a repeat customer upgrade. Total Q4 revenue from these deals: $1,604,800.
This combines Close's deal data with AI-driven analysis. The MCP fetches raw deal records; the AI calculates averages and identifies patterns. Accuracy depends on consistent data entry in Close — missing lead sources or close dates will skew results. Large result sets may hit API rate limits.
Use-case deep-dives
When Close MCP speeds up pre-call research for AEs
A 6-person sales team runs 40+ discovery calls a week. Before each call, reps pull contact history, recent emails, and deal stage from Close. Without the MCP, that's tab-switching between Slack threads, the CRM, and meeting notes. With Close MCP in Switchy, the rep asks "summarize last three touchpoints with Acme Corp" and gets the timeline in-thread. The OAuth2 flow means each rep sees only their own pipeline, so no data leakage across territories. This works best when your Close instance has under 5,000 active leads—larger datasets need scoped queries or you'll wait 8+ seconds per lookup. If your team closes fewer than 10 deals a month, the manual tab is faster than the auth overhead.
Close MCP for post-sale account context transfer
A 3-person CS team inherits accounts from sales the day contracts close. The handoff doc lives in Notion, but deal notes, objections, and stakeholder map live in Close. The CS lead uses Switchy to query "pull all Close activity for accounts closed this week" and pastes the output into the handoff template. This cuts handoff prep from 45 minutes to under 10. The MCP shines here because CS doesn't need write access to Close—they're read-only consumers of sales context. If your Close data includes sensitive commission or forecast fields, confirm the OAuth2 scope excludes them before you grant CS access. Teams with fewer than 20 handoffs a quarter should skip the integration and use a shared Loom.
When Close MCP replaces manual pipeline screenshots
A 4-person leadership team reviews pipeline every Monday at 9am. The VP of Sales used to screenshot Close dashboards and paste them into the slide deck. Now she asks Switchy "show deals over $50k in stage 3 or later" and drops the output into the deck as a table. The OAuth2 token refreshes automatically, so no re-auth dance each week. This saves 15 minutes of prep and ensures the data is current as of meeting start, not Friday afternoon. The trade-off: if your pipeline has custom fields or complex stage logic, the MCP's generic queries might miss nuance. Teams under 10 deals in flight should stick with the native Close UI—it's faster to eyeball than to prompt.
Frequently asked
What does the Close MCP do in Switchy?
The Close MCP connects your Close CRM account to Switchy's AI workspace, letting your team query deal pipelines, contact records, and activity logs through natural language. Your AI agents can pull sales data, check lead statuses, and surface customer context without switching tabs. It uses OAuth2, so Close credentials stay secure and you control which data Switchy can access.
Do I need admin access to connect Close to Switchy?
You'll need a Close account with OAuth app creation permissions, which typically means admin or developer access. During setup, you'll authorise Switchy to read your CRM data through Close's OAuth2 flow. The specific scopes depend on which Close resources your team wants to query — contacts, leads, opportunities, or activities. Check your Close role permissions before starting the connection.
Can the Close MCP update deals or create new contacts?
That depends on which tools the MCP exposes and which OAuth scopes you grant during setup. Some MCPs are read-only by design; others support writes like updating deal stages or logging calls. Since this MCP's tool list isn't finalised yet, assume read access by default and check the connection settings in Switchy to see if write operations are available.
How is this different from using Close's API directly?
The MCP wraps Close's API so your AI agents can query CRM data conversationally, without you writing code or managing tokens. Instead of crafting API requests, you ask questions like "show me deals closing this quarter" and the MCP translates that into Close API calls. It's faster for ad-hoc queries but less flexible than a custom integration if you need complex workflows.
Who on my team should connect the Close MCP?
Whoever owns your Close account and understands which CRM data your team needs in Switchy. Usually that's a sales ops lead or the person who manages integrations. Once connected, any Switchy workspace member can query the data through AI agents, but only the connector can revoke access or adjust OAuth scopes in Close's settings.
More other MCPs
- CloudcartCloudcart
- CloudConvertCloudConvert
- CloudflareCloudflare
- Cloudflare Api KeyCloudflare Api Key
- Cloudflare Browser RenderingCloudflare Browser Rendering
- CloverClover
- CoassembleCoassemble
- CodyCody
- CohereCohere
- Coinbase Agents MCPCoinbase Agents MCP
- Coinbase Wallet MCPCoinbase Wallet MCP
- CoinGeckoCoinGecko