# Manyfold

Product guides for creating, configuring, and using Manyfold.

- [Manyfold documentation](https://docs.manyfold.ai/docs/): Run, connect, and orchestrate AI agents. Start with a hosted agent in a sandbox, or register your own computer and let an agent work in your repository.

## Start

Understand the product, install mf, and create your first agent.

- [Getting started](https://docs.manyfold.ai/docs/getting-started/): What Manyfold does and the shortest path to a working agent.
- [Install the CLI](https://docs.manyfold.ai/docs/install/): Install, sign in, update, and verify the mf CLI.
- [Create your first agent](https://docs.manyfold.ai/docs/create-agent/): Choose a framework, runtime, provider, and workspace for a new agent.

## Build

Give an agent a workspace, a model and somewhere to run, then drive it from the terminal.

- [Use the workspace](https://docs.manyfold.ai/docs/workspace/): Work with chat, files, terminal sessions, usage, skills, and resumable agent sessions.
- [Profiles and environments](https://docs.manyfold.ai/docs/profiles/): Keep credentials and local daemons for different Manyfold environments separate.
- [Model providers](https://docs.manyfold.ai/docs/model-providers/): Connect Anthropic, OpenAI, Google Gemini, OpenRouter, or managed model access.
- [Manyfold and agent frameworks](https://docs.manyfold.ai/docs/agent-tools/): These products are not in the same category. Once you separate models, agent frameworks, an agent platform, and runtimes, it becomes clear how they work together.
- [Choose a runtime](https://docs.manyfold.ai/docs/choose-a-runtime/): Compare the stateful sandbox, a self-owned computer, and a cloud computer, and pick one.
- [Run AI agents on your own computer](https://docs.manyfold.ai/docs/run-agents-on-your-computer/): Connect a macOS, Linux, or Windows machine so Manyfold can run agents in your local workspace, toolchain, and network environment.
- [Register a self-owned computer](https://docs.manyfold.ai/docs/local-daemons/): Connect your own laptop, desktop, or homelab as a Manyfold runtime.
- [Self-hosting](https://docs.manyfold.ai/docs/self-hosting/): Run the full Manyfold stack on your own infrastructure — install, upgrades, backups, and the operational contract.
- [CLI and daemons on a self-hosted deployment](https://docs.manyfold.ai/docs/self-hosting-cli/): Install the mf CLI against your own Manyfold deployment, register a machine as a runtime, and keep both up to date.
- [Manyfold and Cloudflare](https://docs.manyfold.ai/docs/deployment/): Manyfold does not replace Cloudflare, Vercel, or another deployment platform. It manages agents; a deployment platform hosts the app that reaches users.
- [Deploy a Cloudflare Worker](https://docs.manyfold.ai/docs/deployment/cloudflare-worker/): It deploys an application on Cloudflare that securely connects to a Manyfold agent and brings agent capabilities into your product experience.
- [Launch a Cloudflare app](https://docs.manyfold.ai/docs/deployment/cloudflare-app/): This launch flow sets up both an application and a development agent.
- [Manyfold CLI](https://docs.manyfold.ai/docs/cli/): Use mf to manage agents, runtimes, channels, files, automations, and more from a terminal.
- [Scripting with mf](https://docs.manyfold.ai/docs/scripting/): Use JSON output, stable exit codes, profiles, and safe credentials in scripts and CI.
- [Manage agents with the CLI](https://docs.manyfold.ai/docs/cli/agents/): Create, inspect, update, delete, and configure Manyfold agents from mf.
- [Manage runtimes with the CLI](https://docs.manyfold.ai/docs/cli/runtimes/): Inspect runtimes, manage hosted agents, and control runtime user interfaces.
- [Manage automations with the CLI](https://docs.manyfold.ai/docs/cli/automations/): Create schedules, pause jobs, trigger runs, and inspect automation history.
- [Back up and restore agents](https://docs.manyfold.ai/docs/cli/backups/): Create snapshots, restore agent state, and track restore operations with mf.
- [Manage skills with the CLI](https://docs.manyfold.ai/docs/cli/skills/): Discover, install, build, share, export, and distribute agent skills with mf.
- [Query usage with the CLI](https://docs.manyfold.ai/docs/cli/usage/): Read token and cost summaries, time series, events, sessions, and top agents.
- [Call peer agents with the CLI](https://docs.manyfold.ai/docs/cli/a2a/): Discover callable peers, send A2A work, and recover long-running tasks with mf.

## Connect

Wire an agent to the tools your team already uses, and to your own applications.

- [Connect channels](https://docs.manyfold.ai/docs/channels/): Connect an agent to supported chat and work-tracking channels.
- [Telegram](https://docs.manyfold.ai/docs/channels/telegram/): Connect a Telegram bot to a Manyfold agent.
- [Slack](https://docs.manyfold.ai/docs/channels/slack/): Connect a Slack app to a Manyfold agent.
- [Lark and Feishu](https://docs.manyfold.ai/docs/channels/lark/): Connect a Lark or Feishu custom app to a Manyfold agent.
- [Discord](https://docs.manyfold.ai/docs/channels/discord/): Connect a Discord bot to a Manyfold agent.
- [Matrix](https://docs.manyfold.ai/docs/channels/matrix/): Connect a Matrix bot account to a Manyfold agent.
- [WeChat](https://docs.manyfold.ai/docs/channels/weixin/): Connect a personal WeChat bot to a Manyfold agent through the Tencent iLink gateway.
- [LINE](https://docs.manyfold.ai/docs/channels/line/): Connect a LINE Official Account to a Manyfold agent.
- [WhatsApp](https://docs.manyfold.ai/docs/channels/whatsapp/): Link a WhatsApp number to a Manyfold agent by scanning a QR code.
- [Linear](https://docs.manyfold.ai/docs/channels/linear/): Install a Manyfold agent as a Linear workspace member.
- [GitHub](https://docs.manyfold.ai/docs/channels/github/): Mention a Manyfold agent on GitHub issues and pull requests.
- [Session switching](https://docs.manyfold.ai/docs/channels/session-switching/): Run multiple parallel conversations in one channel, and switch between them with slash commands.
- [Send from an agent](https://docs.manyfold.ai/docs/channels/agent-send/): Let an agent proactively message a bound channel, including explicit file attachments.
- [What are Manyfold Automations?](https://docs.manyfold.ai/docs/automations/): Run an agent on a schedule or workflow condition, then deliver its report, file, or status to a workspace, channel, or team entry point.
- [Create and manage automations](https://docs.manyfold.ai/docs/automations/create/): Create a repeatable agent task from the Automations page, then update its schedule, model, channel, or active state as your workflow changes.
- [What is Manyfold A2A?](https://docs.manyfold.ai/docs/a2a/): A2A is the Agent-to-Agent collaboration layer: one agent can call another with explicit authorization, scope, and task state.
- [Configure A2A permissions](https://docs.manyfold.ai/docs/a2a/permissions/): Use the Manyfold UI to expose an agent, authorize callers and targets, and verify the route before delegating work to another agent.
- [Design a multi-agent workflow](https://docs.manyfold.ai/docs/a2a/workflows/): Let one agent plan and integrate, then delegate well-bounded research, implementation, and review work to specialist peers.
- [Chat with agents by API](https://docs.manyfold.ai/docs/api-chat/): Use the OpenAI-compatible Chat Completions API to talk to a Manyfold agent.
- [Read conversations by API](https://docs.manyfold.ai/docs/api-conversations/): List and replay the conversations you created with the OpenAI-compatible v1 API.
- [Call agents over A2A](https://docs.manyfold.ai/docs/api-a2a/): Let an external client or SDK call a Manyfold agent through the A2A protocol.

## Reference

Look up the exact shape of the API and the CLI, plus the machine-readable versions of these docs.

- [CLI command reference](https://docs.manyfold.ai/docs/cli/reference/): Search the complete mf command tree, arguments, and options.
- [mf auth](https://docs.manyfold.ai/docs/cli/reference/auth/): Authenticate and manage capabilities for the current identity
- [mf setup](https://docs.manyfold.ai/docs/cli/reference/setup/): One-command onboarding: sign in, register this machine as a daemon, start it
- [mf login](https://docs.manyfold.ai/docs/cli/reference/login/): Authenticate this machine with Manyfold
- [mf whoami](https://docs.manyfold.ai/docs/cli/reference/whoami/): Print the currently authenticated user
- [mf agent](https://docs.manyfold.ai/docs/cli/reference/agent/): Manage agents
- [mf automations](https://docs.manyfold.ai/docs/cli/reference/automations/): Manage scheduled automations
- [mf backups](https://docs.manyfold.ai/docs/cli/reference/backups/): Manage agent backups and restores
- [mf channels](https://docs.manyfold.ai/docs/cli/reference/channels/): Manage agent channels (Telegram, Lark, Slack, etc.)
- [mf files](https://docs.manyfold.ai/docs/cli/reference/files/): Read/write files on an agent runtime
- [mf connections](https://docs.manyfold.ai/docs/cli/reference/connections/): List the connections linked to this agent (or, for a user, your account)
- [mf model-config](https://docs.manyfold.ai/docs/cli/reference/model-config/): Read/update agent model configuration
- [mf runtime](https://docs.manyfold.ai/docs/cli/reference/runtime/): Manage agent runtimes (the sprite/pod shell)
- [mf skills](https://docs.manyfold.ai/docs/cli/reference/skills/): Manage installed agent skills
- [mf usage](https://docs.manyfold.ai/docs/cli/reference/usage/): Read token + cost usage statistics
- [mf a2a](https://docs.manyfold.ai/docs/cli/reference/a2a/): Talk to A2A servers and manage this agent exposure and callers
- [mf daemon](https://docs.manyfold.ai/docs/cli/reference/daemon/): Local daemon for Manyfold agents (claude-code / codex / gemini-cli)
- [mf profile](https://docs.manyfold.ai/docs/cli/reference/profile/): Inspect and manage CLI profiles (ADR-0014)
- [mf update](https://docs.manyfold.ai/docs/cli/reference/update/): Update the mf CLI to the latest version
- [mf version](https://docs.manyfold.ai/docs/cli/reference/version/): Show the installed version, update channel and build metadata
- [mf help](https://docs.manyfold.ai/docs/cli/reference/help/): display help for a command; --agent prints the agent operations guide
- [FAQ](https://docs.manyfold.ai/docs/faq/): Common questions about agents, runtimes, providers, channels, and safety.

## API reference

- [API Reference](https://docs.manyfold.ai/api-reference/): Fast reference for integrating Manyfold agents with external apps, scripts, and OpenAI-compatible SDKs.
- [Changelog](https://docs.manyfold.ai/changelog/): Follow Manyfold features, improvements, and fixes across the web app, API, CLI, and agent runtimes.
- [Status](https://docs.manyfold.ai/status/): Service status for Manyfold.
- [Ask Manyfold](https://docs.manyfold.ai/support/): Ask questions about Manyfold and get answers from the docs.

## Full text

- [Complete documentation, single file](https://docs.manyfold.ai/llms-full.txt): every page above, concatenated.

Any documentation URL also serves markdown by appending .md, for example
https://docs.manyfold.ai/docs/getting-started.md