What does my AI agent need to connect to SyncNow?

Two things: the SyncNow MCP server, and the SyncNow app open and signed in on the same PC.

When the app starts it registers itself in Claude Code and Codex as an MCP server named syncnow, so with Claude Code or Codex installed on that PC there is nothing to configure.

The agent calls list_projects to see the projects you are a member of, with the conversations of the Conversations tab waiting for your reply on each of them, then open_project with the project's name and its objective, which members see live in the project's Activity; when conversations wait for a reply from your side, its answer starts with ANSWER THESE N CONVERSATIONS FIRST, and the agent is told to answer them before any other work. On its first open it receives SyncNow's operating policy and must acknowledge it; then it reads and changes files with tools such as read_file, grep_files and edit_file, and ends with finish_session, leaving notes for the next agent.

If the app is closed or signed out, the tools answer BRIDGE_UNREACHABLE: someone has to open SyncNow and sign in, and retrying does not help. If you install Claude Code or Codex after SyncNow, restart SyncNow so it can register itself. Both run the SyncNow program with the --mcp argument; for Codex, SyncNow keeps the [mcp_servers.syncnow] table of Codex's config.toml pointing at the current version at every start and leaves the rest of that file as it was. Codex reads that file only when it starts, so after SyncNow has registered itself restart Codex to see the SyncNow tools. Claude Code and Codex are the agents SyncNow sets up by itself.


What happens right after I add a project?
Why can't my agent just use the FTP password?
Getting Started
Projects and the Workspace
Agents and Work Sessions
Planning
Rules
Activity, History and Insights
Managers and Teamwork
Connections, Credentials and Security
Source, Safe Editing and Recovery
Updates, Reports and Troubleshooting