Questions, answered.

Project management for a small team and the AI assistants they already pay for. People use the web app; Claude connects over MCP and works from the same board.

Those track work for people. Projectify is built for people and agents together: claims so they don't collide, a briefing so they don't start cold, and attribution on every change.

The first one claims it. The second is told who holds it and when they were last active, and leaves a comment or asks for the claim.

Small teams, from two people to about twenty, where Claude already does real work: writing code, drafting specs, fixing bugs. If your agents need to know what to work on and what the others are doing, it is for you.

One queue that people and agents both pull from. It hands out the highest-ranked task that is free and not blocked, so nobody has to decide who does what next.

Link a Figma file to a project and say what to look for. An agent reads the file and files what it finds as tasks, each finding once, so re-reading the file never duplicates work.

Yes. Projectify syncs both ways with those boards, so a client can keep theirs while your team and your agents work in Projectify. Conflicts show up on the task and on the board.

Anything that speaks MCP. Claude Code, claude.ai and Claude Desktop connect today with one command or one connector, and each one acts for a named person.

No. An agent acts for a person and shows as "Claude for Noor". You pay for people, not for the assistants they already pay for.

Run claude mcp add --transport http projectify https://api.projectify.si/mcp, then choose Authenticate in /mcp and approve it in your browser.

A lease on a task. The holder is the only one who changes it; every call they make renews the lease, and it frees itself 30 minutes after they go quiet.

One Claude that looks across all your projects, sees what is ready, blocked or waiting for review, and sends the right chat to work on it.

Projectify speaks MCP, so any MCP client can connect. Claude Code, claude.ai and Claude Desktop are the ones we build and test with today.

Every session ends with four short parts: what was done, what is unfinished, what comes next and which tasks it touched. The next session, person or agent, starts from it.

Lasting instructions a person gives ("always…", "never…"). They are saved on the project and shown first in every briefing, so they come before an agent's own defaults.

As short records with the context, the decision, the alternatives considered and the consequences. An agent can propose one; a person accepts it.

Yes. Ask for the claim from the task; the holder is told and can hand it over. If the holder has gone quiet, the claim frees itself after 30 minutes anyway.

Read the brief and rules, find or create tasks, claim them, comment, record decisions, keep docs current and end with a handoff. Every write is checked against the same rules people follow.

Yes. Starter is free. While Projectify is in early access, Team is free too: no card, no trial clock.

Yes. During early access there is nothing to cancel. When paid plans start we will announce prices at least 30 days ahead, and nothing is charged unless you choose a paid plan.

Yes. Projectify is a Node app with Postgres and runs from one Docker Compose file. Or use our cloud at projectify.si.

Yes. Your agents can read everything in your projects over MCP, and if you want a full copy, write to hello@projectify.si and we will send you an export.

In a Postgres database on DigitalOcean servers in London, United Kingdom. Our privacy policy lists everyone who processes data for us.

No. An agent acts for one person and sees exactly what that person can see, at that person's permission level.