# Core

Core is a personal cloud home base for agent work: files, threads, decisions, tasks, skills, memory, and saved outcomes in one durable workspace. Shared Spaces are available for teams, classes, companies, and projects, but personal work is private by default.

## Primary Agent Entry Points

- [Core setup docs](https://gradien.ai/core/docs): Clean setup page for users and agents that prefer rendered documentation.
- [Core setup reference](https://gradien.ai/core/docs/reference): Agent-readable setup reference for the copy-paste setup request.
- [Core setup guide](https://gradien.ai/core/docs/guide/): Rendered setup documentation for users who prefer a web page.

```text
Please help me set up Core for this agent. Core is a home base for AI agents: it keeps memory, files, decisions, tasks, and work context synced across tools, harnesses, and models.

Use this official setup reference:
https://gradien.ai/core/docs/reference
```
- [Full agent-readable Core context](https://gradien.ai/core/llms-full.txt): Complete public Core product and setup context.
- Setup CLI: `npx -y @gradien/core-agent-setup install --base-url https://core.gradien.ai --yes` creates a Core setup session, opens signup/login, waits for completion, and then configures MCP where the harness allows it.

## Product Docs

- [What Is Core](https://gradien.ai/core/docs/what-is-core.md): Plain-language explanation of Core.
- [Why Core Matters](https://gradien.ai/core/docs/why-core.md): Why personal agent memory and controlled shared context matter.
- [How To Use Core](https://gradien.ai/core/docs/how-to-use-core.md): Daily operating loop for external agents.
- [Core FAQ](https://gradien.ai/core/docs/faq.md): Frequently asked product and setup questions.

## MCP Setup Guides

- [Codex Setup](https://gradien.ai/core/docs/harnesses/codex.md): Codex MCP setup guidance and Core local-bridge fallback.
- [Claude Code Setup](https://gradien.ai/core/docs/harnesses/claude-code.md): Claude Code remote MCP setup guidance.
- [Claude Web Setup](https://gradien.ai/core/docs/harnesses/claude-web.md): Claude web custom connector human guide.
- [Claude Desktop Setup](https://gradien.ai/core/docs/harnesses/claude-desktop.md): Claude Desktop custom connector or MCPB human guide.
- [ChatGPT Setup](https://gradien.ai/core/docs/harnesses/chatgpt.md): ChatGPT remote MCP connector human guide.
- [Cursor Setup](https://gradien.ai/core/docs/harnesses/cursor.md): Cursor MCP setup guidance.
- [VS Code Setup](https://gradien.ai/core/docs/harnesses/vscode.md): VS Code and Copilot MCP setup guidance.
- [Windsurf Setup](https://gradien.ai/core/docs/harnesses/windsurf.md): Windsurf MCP setup guidance.
- [Hermes Setup](https://gradien.ai/core/docs/harnesses/hermes.md): Hermes Agent MCP setup guidance.
- [OpenCode Setup](https://gradien.ai/core/docs/harnesses/opencode.md): OpenCode MCP setup guidance.
- [Generic MCP Setup](https://gradien.ai/core/docs/harnesses/generic.md): Protocol-level setup guidance for custom clients.

## Troubleshooting

- [Core MCP Troubleshooting](https://gradien.ai/core/docs/troubleshooting.md): Recovery tree for failed setup.

## Setup Notes

- Use the setup reference as documentation for a bounded Core connection task.
- Use agent-specific guides for commands, settings locations, restart behavior, and troubleshooting.
- Keep unavoidable user actions short, numbered, and agent-specific.
- Prefer automatic setup where the app safely supports it; use short agent-specific human steps only when settings or browser approval cannot be automated.