crmoauth2

Attio

Attio is a fully customizable workspace for your team's relationships and workflows.

Verdict

Attio is a flexible CRM that adapts to how your team actually works. When you @mention Attio in a Space, you can create and update contacts, companies, and deals without leaving the conversation — no more context-switching to log a call or add a note. The AI can search your pipeline, pull deal details, attach notes to records, and surface relationship history on demand. Most useful for sales and customer success teams who want their CRM to stay current without manual data entry. Setup requires OAuth with read/write access to your workspace, so coordinate with whoever manages your Attio account before connecting.

Common use cases

  • Log meeting notes to contacts during debriefs
  • Update deal stages from Slack threads
  • Search pipeline for accounts matching criteria
  • Create new leads from inbound inquiries
  • Pull relationship history before customer calls

Integration

Vendor
Attio
Category
crm
Auth
OAUTH2
Tools
11
Composio slug
attio

Tools

  • Create Note

    This tool creates a new note on a given record in attio. the note can be attached to any record type (like person, company, or deal) and includes a title and content. it requires parameters such as parent object, parent record id, title, an

  • Create Record

    This tool creates a new record in attio for a specified object type (people, companies, deals, users, workspaces, etc.). it requires the object type and a values dictionary containing the attributes for the new record.

  • Delete Note
    destructive

    This tool allows users to delete a specific note in attio by its id. it is implemented via delete https://api.attio.com/v2/notes/{note id} and handles note deletion by validating the provided note id. it complements attio create note functi

  • Delete Record
    destructive

    This tool allows you to delete a record from attio permanently. the deletion is irreversible, and the data will eventually be removed from the system.

  • Find Record

    This tool allows users to find a record in attio by either its unique id or by searching using unique attributes. it provides two methods: one for directly retrieving a record by its id with the get /v2/objects/{object}/records/{record id}

  • Get Object Details

    This tool retrieves detailed information about a specific object type in attio, including all its attributes and their properties. this is useful for understanding what fields are available when creating or updating records of this type.

  • List Lists

    This tool retrieves all lists available in the attio workspace. the lists are returned sorted as they appear in the sidebar. this tool is essential for managing and navigating lists, and is a prerequisite for many list-related operations. i

  • List Notes

    This tool lists all notes associated with a specific record in attio. notes are returned in reverse chronological order (newest first).

  • List Objects

    This tool retrieves a list of all available objects (both system-defined and user-defined) in the attio workspace. it makes a get request to the /v2/objects endpoint and returns a json response containing key metadata about each object, whi

  • List Records

    This tool lists records from a specific object type in attio. it provides simple pagination support and returns records in the order they were created. for complex filtering, use the findrecord action instead. standard object types include:

  • Update Record

    This tool updates an existing record in attio for a specified object type (people, companies, deals, users, workspaces, etc.). it uses patch to partially update only the provided fields, leaving other fields unchanged.

Setup

Setup guide

  1. 11. In your Switchy workspace, navigate to Settings > Integrations and click 'Add Integration'. 2. Select Attio from the CRM category and click 'Connect'. 3. You'll be redirected to Attio's authorization page — sign in with your Attio credentials if prompted. 4. Review the requested permissions (read and write access to records, notes, and lists) and click 'Authorize'. 5. Once redirected back to Switchy, you'll see 'Attio Connected' with a green status indicator. 6. Open any Space and type '@Attio get object details for companies' to test the connection — the AI should return a list of available fields in your company object. 7. To invoke Attio tools during a conversation, @mention Attio and describe what you need (e.g. '@Attio create a note on the Acme Corp record summarizing today's call'). 8. The first time you use a tool, Switchy will confirm the action before executing — after that, it runs automatically unless you've set approval rules.

What teammates see: by default, memories from Attio are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.

Works well with

Top models

Compatibility data appears once enough Spaces have used this MCP together with a given model.

How Switchy teams use it

Not enough Spaces yet to publish anonymised usage stats (we require ≥ 50 Spaces per week).

Starter prompts

Log Call Notes

@Attio create a note on the record for Sarah Chen at Acme Corp. Title it 'Q1 Planning Call' and include these points: discussed budget increase, timeline moved to March, needs pricing for enterprise tier.
Open in a Space →

Find Stalled Deals

@Attio find all deals in the 'Proposal Sent' stage that haven't been updated in the last 14 days. Show me the company name, deal value, and last activity date.
Open in a Space →

Create Lead from Inquiry

@Attio create a new company record for 'Horizon Labs' with website horizonlabs.io, industry 'Healthcare', and a contact named Michael Torres with email mtorres@horizonlabs.io.
Open in a Space →

Prep for Customer Call

@Attio list all notes for the Vertex Systems company record from the past 30 days, then show me the current stage and value of any open deals with them.
Open in a Space →

Update Deal Stage

@Attio find the deal with Cascade Partners and update its stage to 'Negotiation'. Add a note that says 'Legal review in progress, expecting signature by end of week.'
Open in a Space →

Example outputs

Illustrative - representative of the model's voice and quality, not literal recordings.

Prompt

@attio find all companies in our pipeline tagged 'enterprise' and show me their latest activity notes

Output

Found 7 companies tagged 'enterprise' in your Attio workspace. Here's a summary: Acme Corp (last note 3 days ago: 'Demo scheduled for Q1 kickoff'), GlobalTech Industries (last note 1 week ago: 'Waiting on procurement approval'), TechVentures LLC (last note 2 days ago: 'Contract sent, expecting signature this week'). The remaining 4 companies have notes older than 2 weeks. All companies are currently in 'Qualification' or 'Proposal' stages. Would you like me to draft follow-up notes for the companies with older activity?

Notes

This example demonstrates read-heavy operations combining record search with note retrieval. The MCP requires OAuth2 access to your Attio workspace with read permissions on both company records and notes. Response time depends on the number of records being filtered and may take several seconds for large datasets.

Prompt

@attio create a new person record for Sarah Chen at Innovate Labs, mark her as a warm lead, and add a note about our intro call today

Output

Created new person record for Sarah Chen in your Attio workspace. Record details: Email domain linked to Innovate Labs (company record auto-associated), lead temperature set to 'Warm', source marked as 'Referral'. Added note titled 'Intro Call - Jan 2025' with your meeting summary. The record is now visible in your 'Active Leads' list and will appear in today's activity feed. Attio has also triggered your workspace automation to send Sarah to your nurture sequence.

Notes

This showcases the MCP's write capabilities across multiple tools (Create Record, Create Note). Be aware that creating records can trigger workspace automations you've configured in Attio—review your automation rules before bulk operations. The MCP requires write permissions and will fail if required fields for person records aren't provided.

Prompt

@attio review all deals closed this quarter, calculate our average deal size, and draft a summary note for our Q1 board deck

Output

Analyzed 23 deals marked 'Closed Won' in Q1 2025. Average deal size: $47,300 (median: $32,000, range: $8,500–$215,000). Top 3 deals account for 52% of total revenue. Deal velocity averaged 34 days from first contact to close. Here's a draft summary note: 'Q1 2025 Performance: Closed 23 deals totaling $1.09M, exceeding target by 18%. Enterprise segment (>$50K) showed strongest growth with 8 deals. Mid-market conversion rate improved to 31%. Key win: $215K expansion deal with existing customer GlobalTech.' I can create this note in Attio attached to your Q1 workspace record if you'd like.

Notes

This example pairs Attio's data retrieval with AI reasoning to synthesize insights. The MCP fetches deal records and their attributes, then the AI performs calculations and drafting. Accuracy depends on consistent data entry practices in your Attio workspace—inconsistent stage names or missing close dates will skew results. Requires read access to deal records and their custom fields.

Use-case deep-dives

Post-demo follow-up at 6-person sales team

When Attio wins for logging demo notes without leaving Slack

A 6-person sales team runs 20-30 demos a week and needs every rep to log follow-up notes in Attio within an hour of the call. The Create Note and Find Record tools let the team post demo summaries directly from Slack or a shared workspace without opening the Attio web app. OAuth2 auth means each note is attributed to the actual rep, not a bot account. This works cleanly if your Attio object model is stable—if you're still iterating on custom fields or list structures weekly, the Get Object Details tool will lag behind your schema changes and cause friction. If your reps already live in Attio for pipeline work, skip the MCP and train them to use the native app. If they live in Slack or a shared doc tool and Attio is just the system of record, this MCP cuts the context-switch tax in half.

Customer success check-in routing at 10-person team

When this MCP handles quarterly account review prep

A 10-person customer success team runs quarterly business reviews for 80 accounts and needs to pull recent notes, deal history, and contact changes into a shared prep doc. The List Notes and Find Record tools let the team query Attio for the last 90 days of activity on a given company record and paste it into a Google Doc or Notion page without manual export. The List Lists tool helps navigate custom list structures if your Attio workspace uses lists to segment accounts by tier or health score. This breaks down if your QBR prep involves cross-object rollups or calculated fields—Attio's API doesn't expose formulas, so you'll still need to open the web app for revenue trends or churn risk scores. If your QBRs are mostly narrative (notes, emails, meeting summaries), the MCP saves 15 minutes per account. If they're data-heavy, it's a 30% solution.

Investor update data pull at 3-person founding team

When Attio MCP speeds monthly investor reporting

A 3-person founding team sends monthly updates to 12 angel investors and needs to pull pipeline metrics, new customer records, and deal stage changes from Attio into a templated email or slide deck. The Find Record and List Lists tools let the team query specific companies or deals by attribute (like 'closed this month' or 'enterprise tier') and paste the results into their update doc without exporting CSVs. The Delete Record tool is useful for cleaning up test data or duplicate entries before the pull, keeping the output clean. This works if your investor updates are qualitative—names, logos, deal summaries. If you need aggregated metrics (total ARR, conversion rates, pipeline velocity), the MCP can't calculate those; you'll still need Attio's reporting UI or a BI tool. For narrative updates with 10-20 data points, the MCP cuts prep time from 45 minutes to 15.

Frequently asked

What can the Attio MCP do in Switchy?

The Attio MCP lets AI agents read and write your CRM data directly — create people, companies, and deals; add notes to records; search by attributes; and list your workspace objects. Your team can ask questions like 'show me all deals closing this month' or 'add a note to John's record' without switching apps. It uses Attio's full API, so anything you can do in their interface works here.

Do I need admin access to connect Attio via OAuth?

You need workspace member access at minimum, but admin is recommended. Attio's OAuth flow will ask for read and write permissions across records, notes, and lists. If your role restricts certain object types (like deals or custom objects), the MCP won't be able to create or query those. Check your workspace permissions in Attio's settings before connecting.

Can the Attio MCP update existing records or only create new ones?

It can do both. The 'Create Record' tool handles new entries, and you can use 'Find Record' to locate an existing person or company by ID or unique attribute, then update it. The MCP also supports deleting records and notes, though deletions are permanent. If you need bulk updates across hundreds of records, use Attio's CSV import instead.

How does this compare to using Attio's web app or Zapier?

The MCP is faster for ad-hoc queries and multi-step workflows. Instead of clicking through filters or building a Zap for every task, you ask in plain language and the AI chains together searches, creates, and updates. Zapier is better for scheduled automations (like nightly syncs). The web app is still where you configure custom fields and list views.

Who on my team should connect the Attio integration?

Whoever owns your CRM data — usually your sales ops lead or head of sales. That person's Attio permissions determine what the MCP can access for everyone in Switchy. If they leave or their token expires, the integration stops working until someone reconnects. One connection covers your whole Switchy workspace.

Data last verified 607 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.