Typless
Typless is an AI-powered document extraction platform that automates manual data entry from various documents using a simple API.
Verdict
Common use cases
- Check which invoice schemas are live
- Confirm account owner before bulk upload
- Update billing email during handoff
- List document types for new hire onboarding
- Verify API access before automation run
Integration
- Vendor
- Typless
- Category
- docs
- Auth
- API_KEY
- Tools
- 3
- Composio slug
typless
Tools
- Get Document Types
Tool to retrieve a list of all document types available in the account. use when you need to view or select schemas for document processing.
- Get User Profile
Tool to retrieve the authenticated user profile. use after authentication to confirm account details.
- Update User Profile
Tool to update the authenticated user's profile. use when user wants to change first name, last name, or email. example prompt: "update my email to alice@example.com"
Setup
Setup guide
- 11. Open your Switchy workspace and navigate to Settings > Integrations. 2. Find Typless in the MCP directory and click Connect. 3. You'll be prompted to paste an API key — log into your Typless account, go to Settings > API Keys, and generate a new key with read and write permissions. 4. Copy the key and paste it into the Switchy prompt, then click Authorize. 5. Switchy confirms the connection by fetching your user profile in the background. 6. Open any Space and type '@Typless get document types' to verify the integration works — you should see a list of schemas configured in your account. 7. If the command returns an error, double-check the API key has the correct scopes and hasn't expired. 8. Once confirmed, your team can @mention Typless in any Space to query document types or update profile details on the fly.
What teammates see: by default, memories from Typless 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
Starter prompts
List Document Schemas
@Typless show me all document types available in this accountOpen in a Space →
Check My Profile
@Typless get my user profile and show me the current email addressOpen in a Space →
Update Contact Email
@Typless update my email to finance-lead@company.comOpen in a Space →
Verify Schema Readiness
@Typless list document types and confirm we have an 'Invoice' schema configuredOpen in a Space →
Audit Account Owner
@Typless get the user profile and tell me the first and last name on this accountOpen in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Typless what document types are set up in our account? I need to know which schemas we can use for invoice extraction.
Based on your Typless account, you have 4 document types configured: 'US Invoice Standard' (ID: inv_us_001), 'EU VAT Invoice' (ID: inv_eu_002), 'Purchase Order' (ID: po_std_003), and 'Receipt - General' (ID: rcpt_gen_004). The US Invoice Standard schema includes fields for vendor name, invoice number, date, line items, subtotal, tax, and total. This is an illustrative example—your actual document types will reflect your account's custom schemas and training data.
This example shows the MCP's ability to list configured extraction schemas. Useful before processing documents to confirm the right template exists. Requires valid API key authentication. The actual document types returned depend entirely on what schemas your team has created and trained in the Typless dashboard.
@Typless update my profile email to jordan.lee@company.com and change my first name to Jordan
Profile updated successfully. Your email is now jordan.lee@company.com and your first name is set to Jordan. Last name remains unchanged. This is an illustrative example—the actual API response confirms the fields modified and returns the updated user object with a timestamp of the change.
This example demonstrates a write operation that modifies account settings. The MCP can update first name, last name, or email for the authenticated user. Be aware this is a permanent change to your Typless account profile—there's no undo via the MCP. Requires API key with write permissions.
@Typless check my user profile, then suggest which document type I should use for processing vendor invoices from European suppliers
Your Typless profile shows you're authenticated as Alex Chen (alex.chen@company.com), account created March 2024. Based on your available document types, I recommend using the 'EU VAT Invoice' schema (ID: inv_eu_002) for European vendor invoices. This schema is trained to extract VAT numbers, IBAN details, and EU-specific tax fields that US invoice templates don't capture. This is an illustrative example combining profile retrieval with schema recommendation logic.
This example pairs two MCP tools—profile retrieval and document type listing—with the AI's reasoning to provide a contextualized recommendation. Demonstrates how Typless integration works best when the AI synthesizes account state with the user's workflow question. The recommendation quality depends on how well your schemas are named and configured.
Use-case deep-dives
When Typless makes sense for recurring document extraction
A 3-person finance team at a B2B SaaS company processes 200+ vendor invoices monthly. They need line-item data in their ERP but don't want to hire a VA or build custom OCR. Typless wins here if the invoices follow 5-10 consistent templates—its document-type schemas let the team define extraction rules once, then run them in batch. The Get Document Types tool confirms which schemas are live before processing. The trade-off: if your invoices are one-off PDFs with wildly different layouts every time, you'll spend more time tweaking schemas than you save. This MCP is the right call when your document volume justifies setup and your formats repeat.
Borderline fit for support teams handling mixed uploads
A 6-person support team at a fintech startup gets 50 tickets daily, many with uploaded bank statements or ID scans. They want an AI agent to flag which tickets need manual review based on document content. Typless can extract structured data from those uploads if the team pre-defines document types for each category. The Get User Profile tool confirms auth is working before the agent starts pulling data. The problem: support tickets bring unpredictable formats—handwritten notes, screenshots, PDFs with varying quality. If more than 30% of your uploads fall outside your schema library, you'll hit a wall. This MCP works when your attachment types are known and finite, not when every ticket is a surprise.
Strong match for compliance-heavy hiring workflows
A 10-person HR team at a healthcare company onboards 15 new hires monthly, each submitting W-4s, I-9s, and state tax forms. They need extracted data in their HRIS without manual keying. Typless is built for this: the document types are standardized by law, so schemas stay stable across hires. The Update User Profile tool lets the team adjust account settings as compliance rules shift. The constraint is volume—if you're onboarding 200 people a month, you'll want a dedicated compliance platform with deeper audit trails. Below 50 hires monthly, this MCP delivers fast ROI because the documents never change shape and the extraction accuracy compounds over time.
Frequently asked
What does the Typless MCP do in Switchy?
It connects your Typless document extraction account to Switchy's AI workspace. Your team can retrieve document schemas, check account details, and update user profiles without leaving the chat. Useful when you're building workflows that parse invoices, receipts, or forms and need to reference your Typless setup mid-conversation.
Do I need admin access to connect Typless?
You need a Typless API key, which any account holder can generate from their dashboard. No special admin role required. The key authenticates as the user who created it, so profile updates will modify that user's details. If multiple people need access, each should connect their own key.
Can the Typless MCP extract data from documents?
No. This MCP retrieves document type schemas and manages user profiles—it doesn't process or extract data from actual documents. For extraction, you'll still use Typless's API or web interface directly. Think of this as the setup layer, not the processing layer.
Why use this instead of Typless's web dashboard?
Speed. If you're already working in Switchy and need to check which document types are configured or update your email, you skip the context switch. The dashboard is still better for creating new schemas or reviewing extraction results—this MCP handles quick lookups and profile tweaks.
Does connecting Typless count against my Switchy plan limits?
MCP connections don't consume seat licenses, but the AI messages that call Typless tools count toward your team's monthly message quota. A simple profile lookup uses one message; checking document types and updating details in the same thread uses two. Plan accordingly if you're near your limit.