Your ChatGPT subscription can connect directly to Tallyfy through the MCP server at https://mcp.tallyfy.com. Once connected, ChatGPT can search, read, and manage your tasks, processes, and templates using natural language - no API keys to manage and no code to write.
The connection uses OAuth 2.1 with PKCE. ChatGPT handles the entire auth flow when you add the server URL - you sign in with your Tallyfy credentials and authorize access.
Enable Developer Mode in ChatGPT
Click your profile icon in the top-left corner, then go to Settings → Apps and Connectors → Advanced Settings. Enable Developer Mode (required for write access to MCP tools - read-only Deep Research works without it).
Open the Connectors or Apps section
In ChatGPT Settings, navigate to Connectors (or Apps - OpenAI renamed this in December 2025, so the label depends on your version). Click Create or Add Custom App.
Configure the connector
Fill in the connector details:
Name: Tallyfy
Description: Connect to Tallyfy to search and manage workflow tasks, processes, and templates
MCP Server URL: https://mcp.tallyfy.com
Authentication: OAuth 2.1 with PKCE (select this option - ChatGPT will auto-discover the endpoints)
Authorize with your Tallyfy account
Click Connect. ChatGPT discovers the OAuth endpoints by querying https://mcp.tallyfy.com/.well-known/oauth-protected-resource automatically, then redirects you to Tallyfy’s login page. Sign in and click Authorize to grant ChatGPT access to your organization. Refresh tokens are stored securely by OpenAI.
Test the connection
Start a new conversation and try a simple query:
Show me my open tasks in Tallyfy
ChatGPT connects to the MCP server and returns your task list. If you see the tool being called (displayed as a step in the response), the connection is working.
Share with your team (optional)
For team-wide access, publish the connector to your ChatGPT workspace. Admins can control visibility and permissions in workspace settings.
OAuth endpoints reference
ChatGPT auto-discovers these through the /.well-known endpoints - no manual configuration needed. For debugging or custom client setup:
Supported OAuth scopes (ChatGPT requests all of these during the authorization step):
Scope
Access
mcp.tasks.read
Search and retrieve task data
mcp.tasks.write
Create and update tasks
mcp.processes.read
View workflow process status
mcp.processes.write
Manage process lifecycle
mcp.templates.read
Read template configurations
mcp.templates.write
Edit templates, steps, and automations
mcp.forms.read
View form field definitions
mcp.forms.write
Add and update form fields
mcp.users.read
View users and guests
mcp.users.write
Invite and manage users
mcp.automation.read
View automation rules
mcp.automation.write
Create and modify automations
Available MCP tools
Once connected, ChatGPT can use these Tallyfy tools through natural language:
Task tools
get_my_tasks - Get all tasks assigned to the current user (no parameters needed)
get_user_tasks - Get tasks assigned to a specific team member by user ID
get_tasks_for_process - Get all tasks in a specific process by ID or name
create_task_from_text - Create a task from natural language - extracts title, deadline, and assigns to named members or guests
Search tools
search_for_tasks - Search tasks across the organization by keyword
search_for_processes - Search workflow processes/runs by keyword
search_for_templates - Search templates by keyword (use get_all_templates for a full list)
Process tools
get_organization_runs - Get all workflow processes with optional filters for status, template, owner, tags, and type
Template tools
get_template - Get a template by ID or name with full details
get_all_templates - Get all templates in the organization
get_kickoff_fields - Get kick-off form fields for a template
get_step_dependencies - Analyze which automations control when a step appears
assess_template_health - Comprehensive health check covering step clarity, form fields, automations, and deadlines
suggest_step_deadline - Get deadline recommendations for a step
suggest_kickoff_fields - Get suggested kick-off fields based on template content
add_step_to_template - Add a new step to an existing template
add_assignees_to_step - Add member or guest assignees to a step
edit_description_on_step - Update the description/instructions on a step
Form field tools
add_form_field_to_step - Add text, dropdown, date, or other field types to a step
update_form_field - Update an existing form field’s properties
move_form_field - Reorder a form field within a step
delete_form_field - Remove a form field from a step
get_dropdown_options - Get the options for a dropdown field
update_dropdown_options - Update dropdown choices
suggest_form_fields_for_step - Get AI recommendations for useful form fields based on step content
Automation tools
create_automation_rule - Create if-then automation rules on a template
update_automation_rule - Modify an existing automation rule’s conditions or actions
delete_automation_rule - Remove an automation rule
analyze_template_automations - Analyze all automations for conflicts and redundancies
get_step_visibility_conditions - See which automations control a step’s visibility
suggest_automation_consolidation - Get AI recommendations for simplifying automation rules
consolidate_automation_rules - Preview or apply automation consolidation changes
User and guest tools
get_organization_users - Get all organization members with profile data
get_organization_users_list - Get members with minimal data for listing
get_organization_guests - Get guest users with full data
get_organization_guests_list - Get guests with minimal data
invite_user_to_organization - Invite a new member by email
Write operation security
ChatGPT displays explicit confirmation modals before executing any write or modify operation (create, update, delete). This prevents accidental changes and gives you full control over what the AI modifies in your Tallyfy workspace.
Real-world usage examples
Finding urgent tasks
You: "Show me all urgent tasks assigned to the sales team"
ChatGPT: Uses search_for_tasks → returns filtered task list with details
Checking process status
You: "What's the status of the employee onboarding process for John Smith?"
ChatGPT: Uses search_for_processes → returns process details, completion percentage, current step
Template health audit
You: "Analyze our Customer Onboarding template and suggest improvements"
You: "Add a rule to the Sales Template that assigns the contract review step to Sarah when deal value exceeds $50,000"
ChatGPT: Uses get_template to find the step ID → create_automation_rule →
confirms creation after your approval
Cross-referencing data
You: "Find the onboarding template, get all its form fields, then show me
active onboarding processes and highlight which fields are most
commonly left empty"
ChatGPT:
1. get_all_templates to find the onboarding template
2. get_template to retrieve form field details
3. search_for_processes to find active runs
4. Analyzes and summarizes completion patterns
Best practices for natural language queries
Be specific with search terms
Good: "Find all tasks with 'invoice approval' in the title assigned to me"
Bad: "Show tasks"
Provide context for filtering
Good: "Get processes from the Customer Onboarding template created in the last 7 days"
Bad: "Find recent processes"
Specify exactly what you want to modify
Good: "Update the 'Priority' dropdown in the Bug Report template to add 'Critical' as an option"
Bad: "Change the priority field"
Ask for structured summaries
Good: "List all my tasks grouped by due date, showing only title and assignee"
Bad: "What are my tasks?"
Limitations of the text interface
ChatGPT’s text-based UI hits real walls with some Tallyfy features:
Visual workflow representation
ChatGPT cannot display Tallyfy’s visual process tracker or workflow diagrams. You can’t see how steps flow, real-time progress bars, or process dependencies at a glance. For visual process monitoring, use Tallyfy’s native tracker view directly.
Complex form interactions
Dropdown fields with many options, date pickers, and file uploads are difficult in plain text. Multi-select dropdowns become a wall of text; file attachments can’t be handled at all.
Real-time updates
ChatGPT’s turn-based model doesn’t push live updates. Changes made in Tallyfy by teammates won’t appear until you ask again - there are no live notifications.
Bulk visual operations
Reordering many steps via drag-and-drop, watching batch updates propagate, or seeing a template’s visual branching structure - these stay in Tallyfy’s native interface.
Security and data handling
OAuth 2.1 with PKCE: Secure token-based access - no API keys passed around
Granular scopes: 12 separate read/write scopes so you control what ChatGPT can access
Write confirmations: ChatGPT shows explicit modals before any create, update, or delete operation
Token storage: Refresh tokens stored securely by OpenAI, revocable anytime from ChatGPT Settings → Connectors
No training on your data: OpenAI doesn’t use API interactions for model training (per OpenAI’s usage policies)
Prompt injection risk: Be cautious if your task or template content contains prompt-like instructions - verify unexpected ChatGPT behaviors before acting on them
Troubleshooting
“ChatGPT can’t find the Tallyfy connector”
Verify Developer Mode is enabled: ChatGPT Settings → Apps and Connectors → Advanced Settings
Check your ChatGPT plan supports Apps/Connectors (Plus, Pro, Team, Enterprise, Education)
Re-create the connector with the correct URL: https://mcp.tallyfy.com
“Authentication fails or keeps looping”
Re-authorize from ChatGPT Settings → Connectors → Tallyfy → Reconnect
Confirm your Tallyfy account is active and has at least Standard user role
Open https://mcp.tallyfy.com/.well-known/oauth-protected-resource in a browser - a JSON response confirms the server is reachable
Clear browser cache and retry
“ChatGPT says it can’t access certain tools”
Verify you approved all OAuth scopes during the initial connection
Some write operations require Administrator role in Tallyfy
Try disconnecting and reconnecting to re-request all scopes
“Search results are empty or incomplete”
Make queries more specific - search_for_tasks requires a non-empty keyword
Verify the items exist in your Tallyfy workspace
Confirm your Tallyfy user has permission to view those items
“SSE connection errors”
SSE transport is no longer supported - the Tallyfy MCP server uses Streamable HTTP only
Update to the latest ChatGPT version which defaults to Streamable HTTP
Cost considerations
ChatGPT subscription:
ChatGPT Plus: $20/month - MCP via Developer Mode, individual use
All MCP tool calls count toward your Tallyfy API rate limits
Use specific queries to minimize unnecessary tool calls and get the most out of your quota.
Your ChatGPT subscription becomes a workflow intelligence layer - search, analyze, and automate Tallyfy operations using natural language instead of navigating multiple screens.
ChatGPT Enterprise Team and Education users can connect to Tallyfy through MCP servers to manage workflows using natural language with full read/write capabilities through Developer Mode though the text-based interface has significant limitations for visual workflows form interactions real-time collaboration and bulk operations making it best suited for complex searches analysis automation planning and template optimization while still requiring Tallyfy’s native visual interface for process tracking collaboration and interactive form completion.
BYO AI is Tallyfy’s upcoming integration framework that connects existing AI subscriptions like ChatGPT Plus or Claude Pro directly into workflows so AI can automatically complete tasks and generate content without manual copy-pasting between applications.
Tallyfy’s MCP Server enables natural language interaction with workflows through AI assistants by providing tools for searching tasks and processes managing users and templates analyzing workflow health and creating automation rules without requiring API knowledge.
Claude connects to Tallyfy’s MCP server at mcp.tallyfy.com using OAuth 2.1 authentication to give the AI direct access to tasks processes templates and team members through natural language commands in both Claude.ai web and Claude Desktop applications.