Skip to content

MCP server

Tallyfy MCP Server (Preview)

Feature in Development

The Tallyfy MCP Server is currently in development and subject to change. The features and capabilities described here represent the current state of development and may evolve before the official release.

Tallyfy’s MCP (Model Context Protocol) Server lets you control your workflows with plain English through any AI platform that supports MCP. You describe what you want. The AI handles the rest - no API endpoints or complex interfaces required.

What is MCP?

Think of MCP as a universal translator between AI assistants and your business tools. It’s an open standard that creates secure connections, letting AI models talk to external systems while you stay in control of security and permissions.

How Tallyfy MCP Server works

The Tallyfy MCP Server connects AI assistants (Claude, ChatGPT, and others) directly to your Tallyfy organization. Give it your API token. That’s it. Now you can:

  • Ask questions about your processes and tasks in natural language
  • Create and manage tasks through conversation
  • Analyze templates and get suggestions for improvements
  • Search across your organization’s workflows
  • Manage users and guests

Example interactions:

  • “Show me all overdue tasks for John”
  • “Create a task for reviewing the quarterly budget, due next Friday”
  • “What templates do we have for customer onboarding?”
  • “Analyze the health of our ‘Employee Onboarding’ template”

Current capabilities (Preview)

Here’s what the MCP Server can do right now (remember, it’s still in preview):

Search and discovery tools

Search across your organization:

  • search_for_tasks - Find tasks using natural language queries
  • search_for_processes - Locate specific processes by description or criteria
  • search_for_templates - Discover templates that match your needs

Examples:

  • “Find all tasks related to budget approval”
  • “Search for processes containing ‘customer onboarding’”
  • “Show me templates for HR workflows”

User and access management

Organization members:

  • get_organization_users - List all members in your organization
  • get_organization_users_list - Get a simplified list of users
  • invite_user_to_organization - Add new members to your Tallyfy organization

Guest management:

  • get_organization_guests - View all guests with access to your workflows
  • get_organization_guests_list - Get a simplified guest list

Helper functions:

  • resolve_user_ids - Convert user names or emails to internal IDs
  • resolve_group_ids - Convert group names to internal identifiers

Task management and tracking

Personal and team tasks:

  • get_my_tasks - Retrieve your assigned tasks
  • get_user_tasks - View tasks assigned to specific team members
  • get_tasks_for_process - List all tasks within a particular process

Task creation:

  • create_task_from_text - Create new tasks using natural language descriptions with automatic due date extraction

Process monitoring:

  • get_organization_runs - View all running processes with filtering options

Template design and optimization

Template analysis:

  • get_template - Retrieve detailed information about specific templates
  • assess_template_health - Get comprehensive analysis of template effectiveness
  • get_step_dependencies - Understand how steps relate to each other
  • suggest_step_deadline - Get AI-powered deadline recommendations

Form field management:

  • add_form_field_to_step - Add new form fields to template steps
  • update_form_field - Modify existing form field properties
  • move_form_field - Relocate form fields between steps
  • delete_form_field - Remove unnecessary form fields
  • suggest_form_fields_for_step - Get AI suggestions for useful form fields

Dropdown field management:

  • get_dropdown_options - View available options for dropdown fields
  • update_dropdown_options - Modify dropdown choices

Template structure:

  • add_assignees_to_step - Assign members or guests to template steps
  • edit_description_on_step - Update step descriptions and instructions
  • add_step_to_template - Insert new steps into templates

Automation and workflow logic

Automation rules:

  • create_automation_rule - Set up new automations for templates
  • update_automation_rule - Modify existing automation logic
  • delete_automation_rule - Remove automation rules
  • analyze_template_automations - Get insights into current automation setup

Optimization:

  • consolidate_automation_rules - Simplify complex automation setups
  • suggest_automation_consolidation - Get recommendations for automation improvements
  • get_step_visibility_conditions - Understand when steps appear or hide

Kickoff optimization:

  • suggest_kickoff_fields - Get recommendations for process launch form fields

System utilities

Server management:

  • get_server_status - Check MCP server health and connectivity
  • get_available_tools - List all available MCP tools and their descriptions

Security and authentication

The MCP Server needs your Tallyfy API token to connect. Why? Simple:

  • Only authorized users can touch your organization’s data
  • Every action respects your existing permissions
  • Audit trails show exactly who did what
  • Your data stays protected by Tallyfy’s security protocols

Getting started (When available)

Once the MCP Server is officially released, you’ll be able to:

  1. Set up the connection using your Tallyfy API credentials
  2. Configure your AI assistant to connect to the Tallyfy MCP Server
  3. Start conversing with your Tallyfy data using natural language
  4. Perform complex workflows through simple descriptions

Available MCP integrations

Ready to connect? Pick your AI platform:

Future capabilities

Coming soon (we’re working on these right now):

  • Launch and manage processes directly
  • Advanced reports with charts and insights
  • Connect to other middleware platforms
  • Build custom automations just by describing them
  • Real-time team collaboration features

Technical architecture

Under the hood, we’ve built:

  • Core SDK - Python code that talks to Tallyfy’s API
  • MCP Server - The actual protocol implementation
  • Tool Framework - Structured definitions so AI knows what it can do
  • Authentication Layer - Token-based security (your data stays safe)
  • Natural Language Processing - Turns your questions into API calls

Major limitations of text-based AI interfaces

Here’s the thing - MCP is powerful, but text-based AI chat has some frustrating limitations when you’re trying to use Tallyfy’s visual workflow features:

Visual interface constraints

You can’t see what you’re doing. That’s the problem.

  • No visual process tracker: Cannot display Tallyfy’s bird’s-eye view of running processes with visual progress indicators
  • Missing template builder: Cannot show the drag-and-drop template interface or visualize step dependencies and branching logic
  • Lack of aggregated views: Cannot present easily filterable views to see processes and tasks at a glance
  • No real-time progress: Cannot show live updates as team members complete tasks or processes advance

Form field interaction challenges

Remember those nice dropdown menus and date pickers in Tallyfy? In text chat, they turn into a mess:

  • Dropdown fields: When a task has dropdown boxes with multiple options, text-based AI must list all options as plain text, making selection tedious
  • Multi-select inputs: Complex form inputs lose their intuitive interface when reduced to text commands
  • Date/time pickers: Visual calendar and time selection tools must be replaced with manual text entry
  • File uploads: Cannot properly handle file upload fields through conversational interface

Assignment and collaboration barriers

Assigning work to your team? It gets clunky fast:

  • Assignee selection: Cannot easily pick assignees as Tallyfy’s UI suggests members, guests, job titles, and groups with visual indicators
  • Bulk operations: Managing multiple assignments or reassigning many tasks becomes inefficient through individual text commands
  • Team collaboration: Real-time collaborative features like simultaneous template editing are reduced to sequential interactions

Template creation limitations

Want to build a template? Text chat makes it painful:

  • Step visualization: Cannot see all steps in a template at once or understand their relationships
  • Reordering steps: No drag-and-drop capability; must use complex text commands
  • Automation configuration: Complex automation rules are difficult to set up through conversational interfaces
  • Preview functionality: Cannot visually preview how the template will function before deployment

Ideal use cases for MCP with text-based AI

Don’t get us wrong - there are times when text-based MCP absolutely shines. Let’s talk about those.

Natural language search and discovery

This is where MCP really works. Ask complex questions, get specific answers:

  • Template search: “Find all templates related to employee onboarding that include background check steps”
  • Process discovery: “Show me all customer onboarding processes that took longer than 5 days to complete”
  • Task queries: “Which tasks are overdue and assigned to the sales team?”
  • Cross-reference search: “Find templates that use the ‘Budget Approval’ form field”

Intelligent template generation

Got a messy document? AI turns it into clean, structured templates in minutes:

  • Auto-creating form fields: Upload existing forms or documents and let AI automatically generate appropriate form fields with validation rules
  • Flowchart conversion: Study flowcharts or process diagrams to automatically create templates with proper step sequences
  • Automation generation: Analyze business rules in plain language and convert them to automation rules
  • Bulk field creation: Generate multiple related form fields based on document analysis

What-if scenario testing

Want to test changes before making them? MCP lets you run simulations:

  • Automation testing: “If I set up an automation to assign tasks based on deal value, show me how it would route these 5 example deals”
  • Process optimization: “What would happen if we removed the approval step for purchases under $1,000?”
  • Workload analysis: “Show me how task distribution would change if we reassigned all of John’s templates to Sarah”
  • Time estimation: “Based on historical data, estimate completion time if we add a review step”

Intelligent process updates

Your compliance docs changed? AI spots the differences and updates everything in one go:

  • Document-based updates: “Here’s our updated SOX compliance procedure. Update our audit template to match these new requirements”
  • Change highlighting: AI can identify exactly what changed and update only the affected parts
  • Version comparison: Compare old and new process documents to generate precise template modifications
  • Bulk updates: Apply consistent changes across multiple templates based on policy updates

Pattern recognition and optimization

MCP watches how your team actually works - then suggests smart improvements:

  • Ad-hoc task analysis: “Look at one-off tasks added to hiring processes last month and suggest which should be added to the template”
  • Bottleneck identification: “Analyze which steps consistently cause delays and suggest optimizations”
  • Template consolidation: “Find similar templates that could be merged into a single, more flexible template”
  • Best practice extraction: “Identify common patterns in high-performing processes and suggest template improvements”

Complex reporting with citations

Need hard data? AI digs deep and shows you exactly where to look:

  • Performance queries: “Which step in our sales process has the longest average completion time, and which team members are fastest?”
  • Compliance reporting: “Show all instances where required approvals were skipped, with links to specific processes”
  • Trend analysis: “How has our customer onboarding time changed over the last 6 months, and what factors contributed?”
  • Cross-functional insights: “Which departments most frequently add ad-hoc tasks, and what types of tasks are they?”

We’ll let you know the moment this feature goes live. Setup guides are coming too.

Mcp Server > Using Tallyfy MCP server with ChatGPT

ChatGPT Enterprise Team and Education users can now connect to Tallyfy’s MCP server through Deep Research enabling natural language workflow management with powerful search and analysis capabilities while being limited by text-based interactions that lack visual process tracking form field interactions and real-time collaboration features making it best suited as a complement to Tallyfy’s native interface rather than a replacement.

Mcp Server > Using Tallyfy MCP server with Claude (text chat)

Claude Desktop has evolved into the most mature MCP platform with Desktop Extensions for one-click installation Remote MCP servers with OAuth and Claude 4 models achieving 72.5% SWE-bench accuracy enabling powerful natural language workflow automation with Tallyfy while the entire AI ecosystem including OpenAI Microsoft and Google has standardized on MCP creating an explosive growth from 1000 to 7260+ available servers by August 2025.

Mcp Server > Using Tallyfy MCP server with Microsoft Copilot Studio

Microsoft Copilot Studio offers enterprise-grade MCP support for organizations needing AI agents connected to business systems with features like Virtual Network integration Data Loss Prevention multiple authentication methods and multi-agent orchestration capabilities that enable intelligent workflow automation through Tallyfy’s MCP server while requiring additional setup complexity compared to simpler alternatives.

Vendors > OpenAI ChatGPT Agent

OpenAI ChatGPT Agent launched July 17 2025 as an evolution of Operator combining browser automation with deep research code execution and document creation at $200/month for 400 tasks offering 70% better success rates than its predecessor while integrating with Tallyfy through webhooks APIs and MCP servers to orchestrate complex business workflows with natural language control and complete audit trails.
Get in touch
About Tallyfy