Claude integration

Connecting Claude to Tallyfy

Both Claude.ai (web) and Claude Desktop connect to Tallyfy through the MCP server at https://mcp.tallyfy.com. Once connected, Claude gets direct access to your tasks, processes, templates, and team members - letting it read context, take actions, and complete work on your behalf using natural language.

The connection uses OAuth 2.1 with PKCE. Claude handles the entire auth flow when you add the server URL - you just sign in with your Tallyfy credentials and authorize access.

Option 1: Claude.ai web (browser)

Available on paid plans (Pro, Max, Team, Enterprise). Once added, the connector works across Claude web, Claude Desktop, and Claude Mobile automatically.

Open Claude.ai Settings

Go to claude.ai ↗ and click your profile icon in the top-left corner, then select Settings.
Go to Connectors

In Settings, click Connectors in the left sidebar.
Add a custom connector

Click Add custom connector. Enter the following:
- Name: Tallyfy
- URL: https://mcp.tallyfy.com
Then click Save.
Authorize with your Tallyfy account

Claude.ai opens a browser window to Tallyfy’s authorization page. Sign in with your Tallyfy credentials and click Authorize to grant Claude access to your organization.
Verify the connection

Start a new conversation. The Tallyfy connector appears in your active integrations. Try a quick test:
```
What are my open tasks in Tallyfy?
```
Claude calls get_my_tasks and returns your current task list.

Option 2: Claude Desktop app

Available on paid plans (Pro, Max, Team, Enterprise). Download for macOS or Windows.

Install or update Claude Desktop

Download Claude Desktop from claude.ai/download ↗. Make sure you’re on a recent version - remote MCP server support with OAuth requires a current release.
Open Settings

Launch Claude Desktop, then open Settings. On macOS use Cmd+, or click the Claude menu → Settings. On Windows use the gear icon or the application menu.
Go to Integrations

In Settings, navigate to the Integrations tab (sometimes labelled Extensions or MCP servers depending on your version).
Add the Tallyfy MCP server

Click Add integration, Add remote server, or the + button. Enter the Tallyfy MCP server URL:
```
https://mcp.tallyfy.com
```
Claude Desktop automatically discovers the OAuth endpoints - no manual endpoint configuration needed.
Authorize with your Tallyfy account

Claude Desktop opens a browser window to Tallyfy’s authorization page. Sign in with your Tallyfy credentials and click Authorize. The token is stored securely using your OS keychain (macOS Keychain on Mac, Windows Credential Manager on Windows).
Restart Claude Desktop

Quit Claude Desktop completely (Cmd+Q on macOS) and relaunch it. This ensures the MCP connection initializes cleanly on startup.
Verify the connection

Open a new conversation. Look for the tools icon (🔨) at the bottom of the chat input - clicking it shows the available Tallyfy tools. Try a quick test:
```
What are my open tasks in Tallyfy?
```
Claude calls get_my_tasks and returns your current task list.

How Claude integration works

Claude integration leverages Anthropic’s MCP (Model Context Protocol) framework. The Tallyfy MCP server runs on Cloudflare at https://mcp.tallyfy.com and exposes your Tallyfy data as tools Claude can call during a conversation.

The architecture:

Tool discovery: When Claude Desktop connects, it queries the MCP server for the list of available tools
OAuth 2.1 authentication: Claude handles the auth flow automatically - you sign in with your Tallyfy credentials and authorize access
Tool calls: When you ask Claude a question, it calls the appropriate Tallyfy tool with the right parameters
API execution: The MCP server uses your credentials to call the Tallyfy API and returns structured results
Response: Claude formats the data into a natural language answer in your chat

Claude excels at tasks requiring deep analysis, careful reasoning, and nuanced judgment - perfect for complex workflow decisions.

Authentication architecture

The MCP server uses OAuth 2.1 with PKCE (S256). All discovery and token exchange happens through https://mcp.tallyfy.com, which proxies to Tallyfy’s authorization server:

Endpoint	URL
MCP server	`https://mcp.tallyfy.com`
Protected resource metadata	`https://mcp.tallyfy.com/.well-known/oauth-protected-resource`
OpenID configuration	`https://mcp.tallyfy.com/.well-known/openid-configuration`
Authorization (proxy)	`https://mcp.tallyfy.com/mcp/oauth/authorize`
Token exchange (proxy)	`https://mcp.tallyfy.com/mcp/oauth/token`
Client registration (proxy)	`https://mcp.tallyfy.com/mcp/oauth/register`
Tallyfy auth server	`https://account.tallyfy.com`

Supported OAuth scopes:

Scope	Access
`mcp.tasks.read`	Read tasks and assignments
`mcp.tasks.write`	Create and update tasks
`mcp.processes.read`	Read workflow runs
`mcp.processes.write`	Manage processes
`mcp.templates.read`	Read templates and steps
`mcp.templates.write`	Edit templates, steps, and automations
`mcp.forms.read`	Read form fields
`mcp.forms.write`	Add and modify form fields
`mcp.users.read`	Read organization members
`mcp.users.write`	Invite and manage users
`mcp.automation.read`	Read automation rules
`mcp.automation.write`	Create and modify automations

MCP server tools

The Tallyfy MCP server exposes these tools to Claude:

Task tools

get_my_tasks - Get all tasks assigned to the current user (no parameters needed)
get_user_tasks - Get tasks for 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 active 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 based on its type and complexity
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 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 exactly 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

Task automation with Claude’s strengths

Complex document analysis

Claude excels at understanding lengthy, complex documents:

Legal contract review with clause-by-clause analysis
Technical documentation review for accuracy and completeness
Research synthesis from multiple sources
Compliance audit against detailed requirements

Multi-step reasoning

Claude handles tasks requiring careful thought:

Task: Evaluate Project Proposal
Claude instruction: "Review the attached proposal against our evaluation
criteria. Score each section (feasibility, budget, timeline, risk).
Identify any gaps or concerns. Provide a detailed recommendation with
specific conditions for approval."

Writing and documentation

Claude generates high-quality written content:

Task: Create Process Documentation
Claude instruction: "Based on the process steps completed so far, write
detailed documentation including: purpose, prerequisites, step-by-step
steps, common issues, and best practices. Format in our standard
template structure."

Data analysis with context

Claude understands nuanced data patterns:

Task: Analyze Customer Feedback
Claude instruction: "Review all feedback from this quarter. Identify
recurring themes, sentiment patterns, and emerging issues. Correlate
with product changes. Generate actionable insights for each department."

Real-world automation examples

RFP response generation

Workflow: RFP Response Process
Claude's role: Reads RFP requirements, matches against your capabilities
database, generates tailored responses for each section, flags areas
needing human input, maintains consistent tone and messaging.

Employee performance reviews

Workflow: Quarterly Review Process
Claude's role: Synthesizes feedback from multiple sources, identifies
patterns and growth areas, drafts balanced review narrative, suggests
specific development goals, ensures consistency across reviews.

Template health audits

Prompt: "Assess the health of our Customer Onboarding template and
suggest improvements to the automation rules."

Claude calls: assess_template_health → analyze_template_automations →
suggest_automation_consolidation, then summarizes findings and
recommended fixes.

Claude-specific instruction patterns

Leverage Claude’s analytical depth

Good: "Analyze the proposal considering technical feasibility, financial
viability, strategic alignment, and risk factors. Provide weighted scoring
for each dimension with detailed justification."

Bad: "Review the proposal and approve or reject."

Use Claude’s context window effectively

Good: "Using all previous task responses in this process, identify
patterns and generate a complete summary highlighting key decisions
and their rationales."

Bad: "Summarize this process."

Enable structured reasoning

Good: "First, identify all stakeholders affected. Second, analyze impact
on each group. Third, recommend mitigation strategies. Finally, provide
an implementation timeline."

Bad: "Figure out who this affects and what to do."

Advanced Claude features

Long context handling Claude handles up to 200K tokens (1M in beta for some models):

Entire process histories
Multiple document attachments
Full analysis across steps
Pattern recognition over time

Multilingual capabilities Claude processes multiple languages:

Automatic language detection
Translation with context preservation
Cultural nuance awareness
Localized content generation

Performance optimization

Context management strategies:

Send only relevant historical data
Summarize previous steps when possible
Use references instead of full documents
Implement progressive disclosure

Token optimization:

Efficient: "Analyze Section 3.2 of the contract for liability terms"
Wasteful: "Read the entire contract and find liability information"

Response formatting:

Structured: "Return as JSON: {approved: boolean, reason: string, conditions: array}"
Unstructured: "Tell me your thoughts on this"

Security and compliance

Claude integration maintains enterprise security:

No training on your data: Anthropic doesn’t train on API inputs
JWT validation: Tokens validated via RS256 using Tallyfy’s public key - only Tallyfy can issue valid tokens
OS keychain storage: Claude Desktop stores tokens in macOS Keychain or Windows Credential Manager, never in plaintext
Data isolation: Each organization’s data remains separate
Audit logging: Complete interaction history
GDPR compliant: Data deletion on request
SOC 2 Type II: Anthropic maintains compliance

Troubleshooting

“MCP tools icon doesn’t appear”

Restart Claude Desktop completely (Cmd+Q on macOS, not just closing the window)
Check Settings → Integrations to confirm https://mcp.tallyfy.com is listed and active
If the server shows an error, try removing and re-adding it

“Authorization fails or loops”

Make sure you’re signing in with the Tallyfy account that has access to the organization
Try opening https://mcp.tallyfy.com/.well-known/oauth-protected-resource in your browser - if you get a JSON response, the server is reachable
Clear Claude Desktop’s stored credentials and re-authorize from scratch

“Claude times out on complex tasks”

Break into smaller subtasks
Reduce context window size
Implement checkpoints for long analyses

“Responses lack specific details”

Provide more context in your prompts
Include examples of the desired output format
Reference specific data fields by name

“SSE connection errors”

SSE transport is no longer supported - the Tallyfy MCP server uses Streamable HTTP only
Update your Claude Desktop to the latest version, which defaults to Streamable HTTP

Implementation best practices

Start with read-only tools: Use get_my_tasks, search_for_processes, and get_all_templates first to get familiar with the data before running write operations
Provide clear context: Claude performs best with complete information - name the template, process, or user explicitly
Use structured formats: Define expected output schemas for consistent results
Implement review steps: Add human checkpoints for critical decisions like creating automation rules
Monitor and iterate: Review Claude’s outputs to refine your prompts over time

Cost considerations

Using your Claude subscription efficiently:

Claude Pro: $20/month for individual use
Claude Team: $25/user/month with higher limits and admin controls
Claude Max: $100-200/month for higher usage limits
No additional Tallyfy charges for MCP usage
API calls count toward your Anthropic usage quotas

Available Claude models

Tallyfy’s MCP server works with all Claude models through your subscription:

Claude Opus 4.5: Anthropic’s flagship model with the strongest reasoning and agentic capabilities. Best for complex multi-step analysis and long-horizon tasks.
Claude Sonnet 4.5: Best balance of speed and capability with a 1M token context window (beta). Strong coding performance.
Claude Sonnet 4: Fast hybrid model with extended thinking for deeper reasoning when needed.
Claude Haiku 4.5: Fastest and most cost-efficient option, matching Sonnet 4 on many tasks. Good for quick lookups and task creation.

All models support text and image input, multilingual capabilities, and extended reasoning modes.

Claude brings sophisticated analysis to every workflow. Complex decisions that took hours of human review now happen automatically, with reasoning you can trust and audit.

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

Claude Desktop connects to Tallyfy through an MCP server middleware layer enabling natural language workflow management with features like Desktop Extensions for one-click installation and remote OAuth authentication while offering practical capabilities for task management process analysis documentation generation and intelligent routing despite limitations around visual interfaces real-time updates and complex form interactions.

Integrations > BYO AI (Bring Your Own AI)

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.

Integrations > MCP server

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.

Byo Ai > ChatGPT integration

ChatGPT Plus and Pro subscribers can connect to Tallyfy through MCP protocol and OAuth 2.1 authentication to search manage and automate workflow tasks processes and templates using natural language commands directly in their chat interface.

Was this helpful?

Get in touch

About Tallyfy