About StateDock
StateDock is a shared project-state service for teams and coding agents, keeping goals, decisions, and handoffs together. This edition runs in the operator’s Cloudflare account; agents use MCP to read context, pick up ready work, and record handoffs.
Key features
- 01Centralize project goals and decisions
- 02Record handoffs across agents
- 03Deploy in your Cloudflare account
- 04Guides for backup, recovery, and upgrades
Use cases
- 01Restore project context for coding agents
- 02Pass tasks between tools
- 03Record decisions and progress
MCP support
Claude Code, Cursor, and Codex can use MCP to read project context, pick up ready tasks, and write handoff notes. This service is deployed in the operator’s Cloudflare account.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://statedock.dev/mcpclaude mcp add --transport http dev-statedock-statedock https://statedock.dev/mcp{"mcpServers":{"dev-statedock-statedock":{"type":"http","url":"https://statedock.dev/mcp"}}}Open in Cursor{"servers":{"dev-statedock-statedock":{"type":"http","url":"https://statedock.dev/mcp"}}}Open in VS Code{"mcpServers":{"dev-statedock-statedock":{"type":"http","url":"https://statedock.dev/mcp"}}}{"mcpServers":{"dev-statedock-statedock":{"serverUrl":"https://statedock.dev/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEStateDock
Shared project memory for you and your coding agents.
Goals, decisions and handoffs live in one place. Claude Code, Cursor and Codex read context over MCP, pick up ready work and write back handoffs, so the next agent never starts from zero.
This repository is the Cloudflare self-hosted edition. Deploy it in your own Cloudflare account; you operate its Workers, D1, KV, R2, domain, email delivery, and backups. It is separate from StateDock Cloud and has no Cloud subscription tiers. To run StateDock on your own computer or a Linux server with one command, use StateDock Local instead.
- Product and shared feature guide: statedock.dev/doc
- Install and deploy: docs/self-hosting.md
- Back up, restore, and upgrade: docs/operations.md
- Agent and MCP setup: docs/agent-setup.md
- Architecture, validation, and migrations: docs/architecture.md · docs/validation.md · docs/migrations.md
- Security reports: SECURITY.md
Coding-tool plugins
StateDock's separate MIT plugin repository provides shared Skills for saving plans, restoring task context, and recording progress through StateDock Remote MCP. The current packages target the hosted service at https://statedock.dev/mcp; self-hosted deployments can use the Skills with their own MCP setup, but the published packages' default endpoint is the hosted service.
- Codex: available from the plugin repository's local marketplace; installation is documented in its README.
- DeepSeek Harness:
@statedock/dsh-pluginis published for Harness0.2.0-rc.2; see the official community showcase. - ChatGPT app: submitted to OpenAI for review; approval is pending.
- Cursor: Marketplace submission materials are being prepared; installation and approval are not yet verified.
These plugins record and restore project state. They do not execute task code or schedule agents. Plugin code is MIT-licensed; this self-hosted service remains AGPL-3.0-only.
For a complete local install, use the deployment guide above. It covers the required Cloudflare resources, verified email sender, public HTTPS origin, secrets, migrations, deployment, administrator approval, and agent access. The operations guide contains offline backup and recovery steps.
Development checks
Requirements and the repository's checks are listed in CONTRIBUTING.md. Typical verification:
pnpm install --frozen-lockfile
pnpm test
pnpm typecheck
pnpm lint
pnpm format:check
pnpm selfhost:check
License
StateDock is licensed under AGPL-3.0-only. Dependencies retain their own licenses; see THIRD_PARTY_NOTICES.md. The license does not grant rights to StateDock trademarks or imply official endorsement of third-party instances.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.