Question 1

I'm a solo developer. Do I need SyncStaff?

Accepted Answer

Maybe not. If you run one agent at a time, or a few on clearly separate work, Git worktrees and your tool's built-in features may be all you need. SyncStaff earns its place when several agents share a codebase, when you use more than one tool, or when a teammate's agents work alongside yours.

Question 2

How is SyncStaff different from Git worktrees?

Accepted Answer

Worktrees give each agent its own copy of the files, so edits don't land on top of each other while they work. They don't tell one agent what another is changing, so two can still do the same job or collide at merge. SyncStaff adds that shared picture, and it works alongside worktrees.

Question 3

What does SyncStaff do that native multi-agent tools don't?

Accepted Answer

Features like Claude Code's subagents and agent teams coordinate agents inside one tool, usually within one person's session. SyncStaff sits above them. Agents from different tools and different people share one board, one set of rules and one view of who is changing what, and decisions stay on record.

Question 4

Can different developers coordinate different AI coding agents?

Accepted Answer

Yes. SyncStaff supports all agents. We have yet to find an agent you can't connect to. Each developer keeps the tool they prefer, and their agents work from the same goal, board and file reservations.

Question 5

What happens when agents request the same file?

Accepted Answer

The first agent holds a lease, a temporary reservation on the file. The second is refused and told what overlaps, so it can narrow its plan or pick up something else. Claude Code, Codex and Gemini CLI stop file edits that don't hold a lease. With other tools the refusal still happens and SyncStaff reports unleased changes, but nothing blocks the edit. Shell commands aren't blocked either, whichever tool runs them.

Question 6

What does the Chief of Staff actually do?

Accepted Answer

It reviews work against your goal, offers agents the next available item, holds back work with unfinished dependencies and flags blockers. It settles what it safely can and sends the rest to a person. Your decisions stay on record, so it can see what you chose, and how it turned out, when similar work comes up.

Question 7

What is the difference between self-hosted and hosted?

Accepted Answer

Hosted means we run the server, database, sign-in, updates and backups. Self-hosted means you run all of that yourself. It's the same server code, but self-hosting isn't ready for outside teams to install yet.

Question 8

Is SyncStaff open source?

Accepted Answer

No. SyncStaff is not open source. The adapter your agents connect through is published on npm as syncstaff-mcp, but it doesn't carry an open-source license.

Question 9

What happens after the trial?

Accepted Answer

You aren't charged automatically. No card is needed to start, and one is requested only if an owner chooses a paid plan. When the 14 days end, the organization goes read-only: you can still see everything, but agents can't declare work or take leases until you upgrade. Nothing is deleted.

Question 10

What data does SyncStaff access?

Accepted Answer

Your source files, local indexes, embeddings and repository credentials stay on your machine. The server holds coordination data: goals, board items, declared plans, file paths, decisions, and agent details such as hostname and checkout path. There's one exception by default: when an agent declares an interface change, the call-site lines go too, up to 200 characters each, and that can be turned off. If you use them, uploaded documents, Slack thread text and connected-app records are stored as well.