Skip to content
Tallyfy

Find anything across your workflows

Search Tallyfy the way you think

Instead of clicking through screens and setting up filters, just describe what you’re looking for. Your AI searches across tasks, processes, and templates simultaneously and brings back exactly what matches - no hunting required.

What you’ll accomplish

  • Search across tasks, processes, and templates with one natural language question
  • Find items by client name, keyword, template purpose, or any other criteria
  • Skip the click-through-multiple-screens workflow entirely

Try it now

Connect your AI to Tallyfy first, then try any of these prompts directly.

Connect Claude to Tallyfy if you haven’t already, then try:

Find all processes related to Acme Corp
Which template handles vendor onboarding?
Find every open task mentioning "budget approval"

What happens when you ask

When you type a search prompt, your AI figures out what kind of thing you’re looking for and calls the right tool - or combination of tools - to find it.

  • search_for_tasks - keyword search across all tasks in your organization. Good for finding specific action items, task names, or anything assigned to someone.
  • search_for_processes - keyword search across all running processes. Useful when you want to see every active workflow tied to a client, project, or topic.
  • search_for_templates - keyword search across your template library. Helpful when you know roughly what a template does but can’t remember its name.
  • get_all_templates - retrieves your full template list. The AI falls back to this when your search term is broad or when it wants to browse before narrowing down.

For a prompt like “find everything related to Acme Corp”, your AI will likely run search_for_processes and search_for_tasks in parallel. For “which template handles vendor onboarding”, it will call search_for_templates and possibly get_all_templates if the keyword search doesn’t surface a clear match.

You don’t need to think about which tool to use. That’s the point.

How it works behind the scenes

Every search goes through Tallyfy’s MCP server at https://mcp.tallyfy.com. The AI sends your natural language question, picks the appropriate tool or tools, and the server queries your organization’s data using your authenticated credentials.

The flow:

  1. You type a question in plain language
  2. Your AI parses the intent - are you looking for tasks, processes, templates, or some combination?
  3. It calls the relevant search tools with the right keyword parameters
  4. The MCP server queries Tallyfy and returns structured results
  5. Your AI formats those results into a readable response

Results come back with context - not just names, but status, assignees, associated templates, and other details that help you figure out what to do next. If you ask a follow-up question (“who’s the owner of that process?”), the AI can call get_process or get_tasks_for_process to drill deeper.

One search prompt can pull together information that would otherwise require opening three or four different Tallyfy screens. That’s the practical win here.

Tips for better results

Be specific about the subject. “Find the onboarding process for Jane Smith” returns more precise results than “find onboarding”. The more context you give, the less the AI has to guess.

Say what you want returned. “Show me the template name, status, and who owns it” shapes the output so you’re not wading through unnecessary detail.

Combine search types in one prompt. “Find all tasks and processes related to our Q2 audit” works fine - your AI will run both search_for_tasks and search_for_processes at once rather than making you ask twice.

Use purpose or intent, not just names. You don’t need to know the exact template name. “Which template do we use when a new contractor starts?” works just as well as knowing the template is called “Contractor Onboarding v3”.

Add context from other systems. You can paste a client name from your CRM, a project code from a spreadsheet, or a subject line from an email and ask Tallyfy to find anything matching it. The AI handles the translation between how you describe things and how they’re named in Tallyfy.

Narrow down after a broad search. Start with “show me all processes tagged with ‘legal’” and then follow up with “which of those are still active?” rather than trying to construct the perfect query upfront. Iterating is faster than front-loading every condition.

What this looks like in practice

Say you’re jumping into a client meeting in 10 minutes and need to get up to speed quickly. Instead of opening the tracker, filtering by client name, checking the tasks view separately, and then looking up which template is in use - you just ask:

What's the current status of everything we have running for Meridian Financial?
Show me active processes, any overdue tasks, and which templates are involved.

Your AI calls search_for_processes and search_for_tasks simultaneously, combines the results, and gives you a summary. Thirty seconds instead of five minutes of clicking.

Or maybe you’re cleaning up your template library and suspect you have duplicates. Try:

Do we have more than one template for client intake or onboarding?
List all templates with "intake" or "onboarding" in the name or description.

The AI calls get_all_templates or search_for_templates with those keywords and returns a list you can review. You spot three variations of the same process that have accumulated over time - problem found, in a single prompt.

Another common scenario: a colleague is out sick and you need to figure out what they were working on. Rather than requesting admin access to their task queue:

Find all open tasks assigned to sarah@company.com across all active processes

search_for_tasks handles this. You get a list of everything in Sarah’s queue, the processes they belong to, and which ones look urgent based on deadlines.

These aren’t contrived examples. They’re the kinds of lookups people do manually every day - and each one takes a few clicks and some mental overhead that adds up. Having a single place to ask removes that friction entirely.

Take it further

Byo Ai > Use cases

Practical things you can do once your AI is connected to Tallyfy via MCP - from daily task briefings and process status checks to building templates through conversation and setting up automations in plain language.

Use cases > Connect your AI to Tallyfy

One-time setup that works for any major AI platform - Claude, ChatGPT, Copilot, or Gemini - so you can manage tasks and processes and templates through natural language.

Integrations > BYO AI (Bring Your Own AI)

BYO AI lets you connect your existing AI subscriptions like ChatGPT or Claude directly into Tallyfy workflows through the MCP industry standard so your AI can read task context and complete steps and generate content and make decisions automatically inside running processes without any copy-pasting or app-switching.
Get in touch
About Tallyfy