# Connect INTO The Cloud > Connect any MCP client — Claude Desktop, Claude Code, Cursor, or your own — to The Cloud over one endpoint. Auth, the generated tool reference, safety classes, and the consent law, in under 10 minutes. ## Connect INTO The Cloud The Cloud is an **MCP server**. Point any MCP-capable AI at one HTTP endpoint and it reads your memory and acts in your workspace — pages, tasks, calendar, and more — with your consent. No insider setup. ## The endpoint One HTTP endpoint speaking JSON-RPC 2.0. This is the only address you need: ```text https://mcp.thecloud.so/mcp ``` ## Quickstart — connect any client OAuth is the default: your browser opens once to sign in to your Cloud account and the connection is live. No secret is copied, pasted, or stored. 1. **Claude Code** — One terminal command. Claude Code drives the browser sign-in. ```bash claude mcp add --transport http the-cloud https://mcp.thecloud.so/mcp ``` 2. **Cursor** — Add to `~/.cursor/mcp.json` (global) or a project `.cursor/mcp.json`, then Reload Window. First connection opens the browser. ```json { "mcpServers": { "the-cloud": { "url": "https://mcp.thecloud.so/mcp" } } } ``` 3. **Claude Desktop** — Settings → Connectors → Add custom connector. Paste the endpoint and sign in when prompted. ```text https://mcp.thecloud.so/mcp ``` 4. **Any other MCP client** — Add the endpoint as an HTTP MCP server. If the client supports MCP's OAuth flow, sign-in runs on first connect; if it does not, use a Bearer key (below). ## Authentication **OAuth (default). **The browser opens once; nothing is stored on your side. This is the right path for every interactive client. **Static Bearer key. **For CI, headless setups, and clients without MCP OAuth yet (Windsurf, ChatGPT): generate a key under Settings → AI and send it as a Bearer header. ```json { "mcpServers": { "the-cloud": { "type": "http", "url": "https://mcp.thecloud.so/mcp", "headers": { "Authorization": "Bearer " } } } } ``` Key-authentication attempts are throttled to 10 per minute per origin; over budget returns a clean 401. ## What your AI can do Once connected, your assistant reads and writes your Pages and Memory, manages Tasks and Calendar events, tracks Habits, and reaches your Files and Profile — using The Cloud as the memory it carries between sessions. Here is the full surface, by family, generated live from the tool registry: **385 tools across 29 families.** - Memory & Sync (11) — Durable cross-AI memory: read, search, write, and organize what every AI should carry between sessions. - Pages & Spaces (29) — Rich documents and the Spaces holding them — create, read, edit, restructure, tag, and publish. - Tasks & My Day (16) — To-dos, task lists, and the daily page — the user's plan for today and what is outstanding. - Calendar & Booking (15) — Events, availability, free-time search, and public booking types. - Habits (10) — Habit definitions, daily logging, streaks, and statistics. - Files & Drive (11) — Workspace files and folders, plus the searchable index of attached drives. - Tables (18) — MCP-native spreadsheet/database grids — columns, rows, queries, summaries, and sharing. - Forms (16) — Form definitions, fields, capacity, notifications, SEO, and collected responses. - Contacts & CRM (17) — People, contact folders, lead scouting, and drafted outreach to a contact. - Email (18) — Connected Gmail accounts — search, read, send, label, archive, and drafts. - Campaigns & Newsletters (20) — Audience campaigns, deliverable packages, send readiness, newsletters, and CAN-SPAM compliance settings. - Telegram (11) — Connected Telegram chats — search, sync control, quote mining, and contact mapping. - Slack (4) — Ora in the room — read a Space's bridged Slack channels, threads, and history, and post as Ora through the per-class autonomy dial and approval queue. - Instagram (3) — Connection status, posting, and archiving. - Music (6) — The workspace music library and connected Spotify playlists. - Studio Apps (41) — Real multi-file apps built inside the workspace — files, manifests, secrets, rendering, sharing, and verification. - Media & Video (27) — Image and video generation, transcription, conversion, compression, clip styling, blurring, and stream-safety review. - Fonts & Typography (7) — Handwriting-to-font creation from a template photo, uploaded font families, applying a font to an appearance slot, and exporting the packaged files. - Agents & Machines (10) — Dispatch agents to your own attached hardware, watch runs, and collect results. - Loops (9) — Scheduled recurring agent runs — create, customize, start, pause, and run now. - GitHub & Code (13) — Repos, issues, pull requests, code search, and the read/branch/commit path to opening a PR. - Commerce (29) — Storefront catalog, products, variants, inventory, and Amazon product resolution. - Profiles (7) — Public profile pages, link and text blocks, and visibility. - Apps & Assets (4) — The workspace app launcher plus brand-mark and character-asset lookup. - Import & Migration (3) — Bring an existing Notion workspace into The Cloud. - Governance & Consent (7) — The consent floor as tools: record and revoke consent, mint and revoke scoped access, and read the telemetry ledger. - Contribution Ledger (6) — Append-only provenance + credits accounting: register signed contributions against artifacts, walk their lineage, and read the credits they earned. - Workspace Admin (14) — Account provisioning, invites, signup approval, and staged account content. - Whop (3) — The connected Whop account through the generalized credential vault — connection status, allowlisted reads (memberships, companies, plans), and confirm-gated writes under the hardened read-allowlist doctrine. Every tool carries a machine-readable safety class — see the [full tool reference](/docs/tool-reference) and [safety classes & the consent law](/docs/safety-and-consent). ## Connecting your own service as a tool The reverse arrow — making your MCP server usable inside The Cloud — enters through the curated Tools catalog today, and the open connect-any-server tier once containment ships. The gate is stated honestly on the [Open connect tier](/docs/open-connect-tier) page. ## Troubleshooting - **The connector will not connect. **Confirm you pasted the exact endpoint. Remove the connector and add it again — the browser sign-in runs once more and the session refreshes. - **An API-key client returns 401. **The key is missing or was revoked. Generate a fresh one under Settings → AI and update the Authorization header. - **Tools look stale after a change. **Reconnect the server, or restart your client, so it re-reads the tool list.