
Utter
Utter L.L.C-FZ · Productivity
Utter is a project management platform for teams and AI agents, now branded as Taskfolk. It puts boards, backlogs, sprints, timelines, docs, goals, and reports in one workspace, then lets you hand certain issues to AI teammates that connect over MCP or the REST API. Your people stay in charge. Agents pick up extra capacity when a task suits them, and the work stays visible on the same board.

About Utter
What Is Utter
Utter is a project management platform built for teams that also work with AI agents. The core job is familiar. Plan work on boards and backlogs, run sprints, map timelines, write docs, and report on delivery without spreading the project across five different tools. So what makes it different? The same workspace treats an AI agent as a teammate you can assign work to.
The product is now sold under the name Taskfolk, run by Utter L.L.C-FZ in Dubai. The pitch is straightforward. You keep a single plan from the first brief to the final handoff, and when an issue suits an agent, you delegate it with a clear scope. The brief, the run, the files, the decisions, and the review all stay attached to that issue, so an agent's change reads the same as a person's.
The biggest limitation is the shape of the AI side. Utter doesn't run its own model. Agents are bring-your-own: you connect Claude Code, Codex, Cursor, or any MCP client, and the platform orchestrates and records the work. If you expect a built-in chatbot that does everything on its own, that isn't what this is. I'm not sure how mature some of the agent handoff features feel in daily use yet, since the workflow depends a lot on the external tool you connect.
Getting Started
- Create a workspace at taskfolk.ai with a magic link. There's no password to set up, and your first workspace starts on a full 14-day Pro trial with no credit card.
- Invite your team. Members and free viewers are unlimited on every plan, and you only pay for builders who create and move work.
- Set up your first project and pick a view. Boards, backlogs, lists, calendars, and timelines all read from the same issues, so you choose what fits the moment.
- Connect an agent if you want one. Mint an API key or add the MCP server, then point Claude Code, Cursor, or another client at your workspace.
- Assign the first agent task with a clear scope and a definition of done, then review the result before it moves on.
Product Information
A quick look at Utter's pricing, supported platforms, and performance.
Best for
The users, tasks, and scenarios where this tool fits best.
Users
- Product and software teams
- Marketing and creative teams
- Operations teams
- Team leads and leadership
- Developers building integrations
Tasks
- Break a PRD into a backlog
- Run a two-week sprint
- Prepare a status update
- Delegate a well-scoped issue to an agent
- Import from another tool
Scenarios
- A small product team that wants Jira-style planning but pays per builder instead of per seat, so clients and part-time reviewers stay free.
- A mixed team where designers, engineers, and agents all touch the same project and everyone needs to see what changed and who changed it.
- A developer who wants Claude Code or Cursor to read the right project context and make traceable changes without a separate internal tool.
- An operations group tracking recurring work and handoffs, where forms and repeating tasks replace chasing people over chat.
- A startup that needs five full projects free, with unlimited issues, before it decides whether to upgrade.
Key features
Boards, Backlogs, and Sprints in One Place
The core planning surface covers summary, board, backlog, list, calendar, sprints, timeline, and docs views, all reading from the same set of issues. Move an issue once and every view updates. Most teams lose hours reconciling a roadmap deck against a task tracker. Here there's a single source to reconcile against, which is the point of an AI task manager that also tracks plain human work.
Bring-Your-Own AI Agents Over MCP
Utter connects the coding agents your team already uses, including Claude Code, Codex, Cursor, and any MCP-compatible client. The platform orchestrates and records what an agent does rather than running a model itself, so you're not locked into one vendor's assistant. Agents get real tasks with scope, and the run stays on the issue. Teams that want MCP project management without building their own plumbing get the connection for free.
Traceable Agent Actions
Every agent-created comment, status transition, and issue update lands in activity history alongside human actions. A change made by an agent reads the same as one made by a person, with a clear record of who did what. No black box. That's what keeps trust intact when automation does real work, and it's the feature most teams end up relying on.
AI Work Breakdown From a Brief
The assistant reads a PRD, FRD, spec, or a rough brief and drafts a structured backlog: epics, the stories and tasks under each, and subtasks under those. It asks clarifying questions when the brief is ambiguous, and you review and edit the tree before anything is created. Nothing gets built behind your back. It turns a messy document into something you can start planning against.
Project-Aware Assistant
Ask about sprint risk, blocked issues, release scope, or recent comments, and the assistant answers using real workspace context. You don't paste in a status dump. It also drafts issues from a short description, filling in a title, type, and body you can edit before saving. That last part saves real time on a messy Monday morning.
REST API, OpenAPI, and Webhooks
The developer surface is a versioned REST API of roughly 180 operations covering issues, projects, sprints, comments, and agents, with a full OpenAPI contract you can generate clients from. If you want REST API project management that behaves predictably in production, this is the layer to build on. API keys are scoped to the permissions each integration needs, and webhooks subscribe external systems to issue, comment, project, and workspace changes with retryable delivery. Usage and audit views show every call.
Free Viewers, Paid Builders
The pricing model charges only for builders. Those are the people who actively create and manage work. Viewers who read projects and follow issues are free on every plan with no cap. For teams that invite clients and part-time reviewers, this changes the math considerably compared with per-seat tools that bill every person.
Reports, Goals, and Project Health
Track goals, workload, milestones, and project health without maintaining a second plan. Reports and full activity history sit alongside the work, so managers can follow delivery without asking for a separate update. That's the practical win of keeping planning and reporting in the same workspace.
Pros and cons
Pros
- Every feature is on the free plan, including the REST API, MCP server, and named AI agents, so you can evaluate the full product before paying. The ceiling is capacity, not features.
- Free viewers with no cap make it cheap to bring in clients and managers who only need to read.
- Agent actions are recorded in the same activity history as human actions, which keeps AI work reviewable instead of hidden.
- Builder seats start at $3 per month and top out at $6, well below the per-seat standard plans of Jira, Linear, ClickUp, and Asana.
- One workspace covers boards, backlogs, sprints, timelines, docs, goals, and reports, so there's less tool-switching.
- Bringing your own agent over MCP or the API avoids locking you into a single vendor's assistant.
Cons
- Utter doesn't run its own model, so the AI experience depends entirely on the external agent you connect. There's no built-in assistant that works on its own, which can frustrate people expecting a turnkey chatbot.
- The free plan caps you at 5 projects and 128 MB of storage, so growing teams hit a ceiling quickly. You keep unlimited issues, but the project count is the real wall.
- Agent features lean on MCP and API setup, which takes some configuration. Teams without a developer will find the agent side harder to adopt.
- The product is rebranding from Utter to Taskfolk, so the name you search for and the name on the site may not match yet, which can confuse new users.
Frequently asked questions
Utter is a project management platform for teams and AI agents. It handles boards, backlogs, sprints, timelines, docs, goals, and reports in one workspace, and lets you delegate well-scoped issues to AI agents connected over MCP or the REST API.
Related content
Explore related tools, skills, and articles for Utter.
Utter Alternatives
Base Code
Base44 (Wix) · Coding · ProductivityBase Code is a shared cloud development environment from Base44. Teammates describe changes to an existing GitHub codebase in chat, preview them live and send them back as pull requests.
Knocket
Knocket (Tencent RTC team) · Productivity · Business · ChatbotKnocket is a free live chat and contact page tool that lets visitors message you from your website, app, or a shareable link, with every reply handled from Telegram or one inbox.
FloorGen
Vinit Kumar (FloorGen) · Productivity · China AIFloorGen is a browser-based floor plan maker that turns a sentence or a sketch into an editable 2D plan and a 3D walkthrough. Export to PDF, DXF or video, and draw without an account.
