OpenClaw 가이드

Automated, hardened OpenClaw installation with Ansible, Tailscale VPN, and firewall isolation

Run OpenClaw Gateway 24/7 on an Azure Linux VM with durable state

Bun workflow for installs and package scripts; Node is required at runtime

ClawDock shell helpers for Docker-based OpenClaw installs

Stable, extended-stable, beta, and dev channels: semantics, switching, pinning, and tagging

Host OpenClaw on a DigitalOcean Droplet

Optional Docker-based setup and onboarding for OpenClaw

Shared Docker VM runtime steps for long-lived OpenClaw Gateway hosts

Run OpenClaw Gateway on exe.dev (VM + HTTPS proxy) for remote access

Step-by-step Fly.io deployment for OpenClaw with persistent storage and HTTPS

Run OpenClaw Gateway 24/7 on a GCP Compute Engine VM (Docker) with durable state

Run OpenClaw Gateway 24/7 on a cheap Hetzner VPS (Docker) with durable state and baked-in binaries

Host OpenClaw on Hostinger

How the installer scripts work (install.sh, install-cli.sh, install.ps1), flags, and automation

Deploy OpenClaw Gateway to a Kubernetes cluster with Kustomize

Run OpenClaw in a sandboxed macOS VM (local or hosted) when you need isolation or iMessage

Install OpenClaw declaratively with Nix

Deploy OpenClaw on Northflank with one-click template

Host OpenClaw on Oracle Cloud's Always Free ARM tier

Run OpenClaw in a rootless Podman container

Deploy OpenClaw on Railway with one-click template

Host OpenClaw on a Raspberry Pi for always-on self-hosting

Deploy OpenClaw on Render with Infrastructure-as-Code

Uninstall OpenClaw completely (CLI, service, state, workspace)

Host OpenClaw on Upstash Box with keep-alive and SSH tunnel access

Agent bootstrapping ritual that seeds the workspace and identity files

Curated links to the most used OpenClaw docs.

Hubs that link to every OpenClaw doc

Backstory and lore of OpenClaw for context and tone

First-run setup flow for OpenClaw (macOS app)

Overview of OpenClaw onboarding options and flows

Implementation plan for the custodian onboarding redesign (living document)

End-to-end guide for running OpenClaw as a personal assistant with safety cautions

Quick start has moved to Getting Started.

Advanced setup and development workflows for OpenClaw

Community-built projects and integrations powered by OpenClaw

CLI onboarding: verify inference, then hand remaining setup to OpenClaw

Scripted onboarding and agent setup for the OpenClaw CLI

Step-by-step behavior for openclaw onboard: what each step does, config it writes, and internals

Migration hub: cross-system imports, machine-to-machine moves, and plugin upgrades

Move Claude Code and Claude Desktop local state into OpenClaw with a previewed import

Move from Hermes to OpenClaw with a previewed, reversible import

Broadcast a WhatsApp message to multiple agents

Launch self-contained OpenClaw HTML widgets inside Discord Activities

Group chat behavior across surfaces (Discord/iMessage/Matrix/Microsoft Teams/QQBot/Signal/Slack/Telegram/WhatsApp/Zalo)

Channel location parsing and portable outbound location payloads

How OpenClaw upgrades the previous Matrix plugin in place, including encrypted-state recovery limits and manual recovery steps.

Matrix MessagePresentation metadata for OpenClaw-aware clients

Per-recipient Matrix push rules for quiet finalized preview edits

Let supported group rooms provide quiet context unless the agent sends with the message tool

WhatsApp group message handling — activation, allowlists, sessions, and context injection

Synthetic Slack-class channel plugin for deterministic OpenClaw QA scenarios

WeChat channel setup through the external openclaw-weixin plugin

Yuanbao bot overview, features, and configuration

Zalo ClawBot channel setup through the external openclaw-zaloclawbot plugin

Zalo personal account support via native zca-js (QR login), capabilities, and configuration

Redirect to /gateway/authentication

Redirect to Task Flow

Redirect to /automation

Redirect to /automation/cron-jobs

Redirect to /cli/message

Redirect to /automation/cron-jobs

Redirect to /automation/cron-jobs

How OpenClaw summarizes long conversations to stay within model limits

Context: what the model sees, how it is built, and how to inspect it

Context engine: pluggable context assembly, compaction, and subagent lifecycle

One rolling conversation across all your channels: the personal-agent default

Markdown formatting pipeline for outbound channels

The default SQLite-based memory backend with keyword, vector, and hybrid search

AI-native cross-session memory via the Honcho plugin

Local-first search sidecar with BM25, vectors, reranking, and query expansion

How memory search finds relevant notes using embeddings and hybrid retrieval

Message flow, sessions, queueing, and reasoning visibility

How OpenClaw presence entries are produced, merged, and displayed

Trimming old tool results to keep context lean and caching efficient

Search past session transcripts and reopen the matching context

Durable session state signal log: state versions, watchers, stale-state notices, and reconciliation

Agent tools for cross-session status, recall, messaging, and sub-agent orchestration

Use SOUL.md to give your OpenClaw agent an actual voice instead of generic assistant sludge

Where timezones show up in OpenClaw — envelopes, tool payloads, system prompt

Public REST API (v1) overview and conventions.

How to submit a copyright or other content rights concern about material published on ClawHub.

How ClawHub listings, versions, installs, publishing, scans, and updates work.

HTTP API reference (public + CLI endpoints + auth).

How ClawHub reports, moderation holds, hidden listings, bans, and account standing work.

How to request ClawHub review for org, brand, owner-handle, package-scope, skill-slug, or namespace ownership disputes.

Fix ClawHub plugin package validation findings before publishing

Skill folder format, required files, supporting artifacts, limits.

Agent loop lifecycle, streams, and wait semantics

WebSocket gateway architecture, components, and client flows

Move one OpenClaw session's reply route between linked chat channels

Status and cleanup guidance for retired inferred follow-up commitments

Delegate architecture: running OpenClaw as a named agent on behalf of an organization

What experimental flags mean in OpenClaw and which ones are currently documented

OpenClaw capabilities across channels, routing, media, and UX.

Mantis captures visual end-to-end evidence for live transport comparisons and focused candidate-only browser proofs, then attaches the artifacts to PRs.

Operator runbook for Mantis Slack desktop QA: GitHub dispatch, local CLI, warm VNC leases, hydrate modes, timing interpretation, artifacts, and failure handling.

Status of the durable message receive/send lifecycle: what shipped, what changed from the original design, and what remains open

Multi-agent routing: agent boundaries, channel accounts, and bindings

How session ownership and presence work when several people operate one agent

OAuth in OpenClaw: token exchange, storage, and multi-account patterns

Run parallel specialist agents without clogging shared model and tool capacity

Local qa-channel scenarios for privacy-preserving personal assistant workflow checks.

Progress drafts: one visible work-in-progress message that updates while an agent runs

QA stack overview: qa-lab, qa-channel, repo-backed scenarios, live transport lanes, transport adapters, and reporting.

Auto-reply queue modes, defaults, and per-session overrides

How active-run steering queues messages at runtime boundaries

Retry policy for outbound provider calls

Streaming + chunking behavior (block replies, channel preview streaming, mode mapping)

What the OpenClaw system prompt contains and how it is assembled

TypeBox schemas as the single source of truth for the gateway protocol

When OpenClaw shows typing indicators and how to tune them

Usage tracking surfaces and credential requirements

Run agent tasks in isolated git checkouts with automatic snapshots and cleanup

Resolve Gateway secrets with the 1Password CLI and let agents use the bundled 1password skill

Metadata-only audit history for agent runs, tool actions, and opt-in message lifecycles

Model authentication: OAuth, API keys, Claude CLI reuse, and Anthropic setup-token

Background exec execution and process management

Bonjour/mDNS discovery + debugging (Gateway beacons, clients, and common failure modes)

Historical bridge protocol (legacy nodes): TCP JSONL, pairing, scoped RPC

CLI backends: local AI CLI fallback with optional MCP tool bridge

Build a third-party operator or WebChat client for the Gateway WebSocket protocol

Agent defaults, multi-agent routing, session, messages, and talk config

Tools config (policy, experimental toggles, provider-backed tools) and custom provider/base-URL setup

Schema-accurate configuration examples for common OpenClaw setups

Node discovery and transports (Bonjour, Tailscale, SSH) for finding the gateway

Doctor command: health checks, config migrations, and repair steps

Supervise the OpenClaw Gateway as a child process from Electron or another host app

Current integration path for external apps, scripts, dashboards, CI jobs, and IDE extensions

Gateway singleton guard: file lock plus WebSocket/HTTP bind

Start local model servers on demand before OpenClaw model and embedding requests

Logging surfaces, file logs, WS log styles, and console formatting

Host multiple tenant trust domains as one isolated OpenClaw Gateway cell per tenant

Run multiple OpenClaw Gateways on one host (isolation, ports, and profiles)

Redirect to /network#core-model

Expose an OpenAI-compatible /v1/chat/completions HTTP endpoint from the Gateway

Expose an OpenResponses-compatible /v1/responses HTTP endpoint from the Gateway

Use OpenShell as a managed sandbox backend for OpenClaw agents

Export OpenClaw diagnostics to OpenTelemetry collectors or stdout JSONL via the diagnostics-otel plugin

Operator roles, scopes, and approval-time checks for Gateway clients

Node capability approvals: how nodes gain command exposure after device pairing

Expose OpenClaw diagnostics as Prometheus text metrics through the diagnostics-prometheus plugin

Gateway WebSocket protocol: handshake, frames, versioning

Remote access using Gateway WS, SSH tunnels, and tailnets

SSH tunnel setup for OpenClaw.app connecting to a remote gateway

What survives a gateway restart or crash: interrupted agent turns resume automatically, subagents and background tasks recover, queued deliveries drain

Why a tool is blocked: sandbox runtime, tool allow/deny policy, and elevated exec gates

How OpenClaw sandboxing works: modes, scopes, workspace access, and images

Contract for `secrets apply` plans: target validation, path matching, and `auth-profiles.json` target scope

Reference for every Gateway rate limit: pre-auth lockouts, browser and webhook throttles, the control-plane write backstop, ACP session caps, and restart cooldown

How OpenClaw handles local file access safely, and why the optional fs-safe Python helper is off by default

Plain-English and technical explanation of npm shrinkwrap in OpenClaw releases

Integrated Tailscale Serve/Funnel for the Gateway dashboard

Invoke a single tool directly via the Gateway HTTP endpoint

Deep troubleshooting runbook for gateway, channels, automation, nodes, and browser

Delegate gateway authentication to a trusted reverse proxy (Pomerium, Caddy, nginx + OAuth)

How inbound audio/voice notes are downloaded, transcribed, and injected into replies

Camera capture on iOS, Android, macOS, and Linux nodes for photos and short video clips

Capability-based desktop control through the computer tool and computer.act node command

Image and media handling rules for send, gateway, and agent replies

Location command for nodes, platform permission modes, and Linux GeoClue setup

Inbound image/audio/video understanding (optional) with provider + CLI fallbacks

Detect the Mac you most recently used and route node alerts there

Talk mode: continuous speech conversations across local STT/TTS and realtime voice

Troubleshoot node pairing, foreground requirements, permissions, and tool failures

Global voice wake words (Gateway-owned) and how they sync across nodes

Android app (node): connection runbook + Connect/Chat/OpenClaw/Voice/Canvas command surface

Redirect to /install/digitalocean

Run the OpenClaw Gateway on EasyRunner with Podman and Caddy

iOS node app: connect to the Gateway, pairing, canvas, and troubleshooting

Enable and invoke privacy-gated HealthKit summaries from an iOS node

Linux support + companion app status

Gateway runtime on macOS (external launchd service)

Agent-controlled Canvas panel embedded via WKWebView + custom URL scheme

Gateway lifecycle on macOS (launchd)

Setup guide for developers working on the OpenClaw macOS app

How the macOS app reports gateway/channel health states

Menu bar icon states and animations for OpenClaw on macOS

OpenClaw logging: rolling diagnostics file log + unified log privacy flags

Menu bar status logic and what is surfaced to users

PeekabooBridge integration for macOS UI automation

macOS permission persistence (TCC) and signing requirements

macOS app flow for controlling a remote OpenClaw gateway

Signing steps for macOS debug builds generated by packaging scripts

macOS Skills settings UI and gateway-backed status

Voice overlay lifecycle when wake-word and push-to-talk overlap

Voice wake and push-to-talk modes plus routing details in the mac app

How the mac app embeds the gateway WebChat and how to debug it

macOS IPC architecture for OpenClaw app, gateway node transport, and PeekabooBridge

Install and use the OpenClaw macOS menu bar app

Redirect to /install/oracle

Redirect to /install/raspberry-pi

Windows support: Windows Hub, native CLI and Gateway, WSL2 gateway setup, node mode, and troubleshooting

Contributor guide for adding a new shared capability to the OpenClaw plugin system

Expose selected Gateway control-plane methods through the bundled, opt-in admin-http-rpc plugin

Redirects to Building Plugins (registering tools section)

Plugin internals: capability model, ownership, contracts, load pipeline, and runtime helpers

Plugin architecture internals: load pipeline, registry, runtime hooks, HTTP routes, and reference tables

Redirects to the current Building Plugins guide

Create your first OpenClaw plugin in minutes

Install and use Codex, Claude, and Cursor bundles as OpenClaw plugins

Build a plugin that registers a local AI CLI backend

Set up Codex Computer Use for Codex-mode OpenClaw agents

Configuration, auth, discovery, and app-server reference for the Codex harness

Runtime boundaries, hooks, tools, permissions, and diagnostics for the Codex harness

Configure native Codex plugins for Codex-mode OpenClaw agents

Browse non-archived native Codex sessions and paginated transcripts across OpenClaw nodes

Find and publish community-maintained OpenClaw plugins

Plugin compatibility contracts, deprecation metadata, and migration expectations

How OpenClaw installs plugin packages and resolves plugin dependencies

Google Meet plugin: join explicit Meet URLs through Chrome or Twilio with agent talk-back defaults

Test packaged plugin overrides with setup-time install flows

Run local GGUF text inference and memory embeddings in OpenClaw with llama.cpp

Optional automatic work journal built from periodic screen snapshots

Plugin manifest + JSON schema requirements (strict config validation)

Choose and configure Google Meet, Microsoft Teams, or Zoom meeting participation

Configure the official external LanceDB memory plugin, including local Ollama-compatible embeddings

memory-wiki: compiled knowledge vault with provenance, claims, dashboards, and bridge mode

Semantic message cards, charts, tables, controls, fallback text, and delivery hints for channel plugins

Bundled `oc-path` plugin: ships the `openclaw path` CLI for the `oc://` workspace-file addressing scheme

Use the optional 1Password plugin as an audited agent secrets broker

Generated inventory of OpenClaw plugins shipped in core, published externally, or kept source-only

Ask users to approve plugin tool calls and plugin-owned permission prompts

Generated index of OpenClaw plugin reference pages

Experimental SDK surface for plugins that replace the low level embedded agent executor

Inbound event helpers for channel plugins: context building, shared runner orchestration, session record, and prepared reply dispatch

Experimental channel ingress API for inbound message authorization

Redirect to /plugins/sdk-channel-outbound

Outbound message lifecycle API for channel plugins: adapters, receipts, durable sends, live preview, and reply pipeline helpers

Step-by-step guide to building a messaging channel plugin for OpenClaw

Redirect to /plugins/sdk-channel-inbound

Reference for defineToolPlugin, definePluginEntry, defineChannelPluginEntry, and defineSetupPluginEntry

Migrate from the legacy backwards-compatibility layer to the modern plugin SDK

Import map, registration API reference, and SDK architecture

Step-by-step guide to building a model provider plugin for OpenClaw

api.runtime -- the injected runtime helpers available to plugins

Setup wizards, setup-entry.ts, config schemas, and package.json metadata

Plugin SDK subpath catalog: which imports live where, grouped by area

Testing utilities and patterns for OpenClaw plugins

Microsoft Teams meetings plugin: join work or consumer meetings as a Chrome browser guest

Build simple typed agent tools with defineToolPlugin and openclaw plugins init/build/validate

Use the bundled Vault plugin to resolve SecretRefs from HashiCorp Vault

Place outbound and accept inbound voice calls via Twilio, Telnyx, or Plivo, with optional realtime voice and streaming transcription

Webhooks plugin: authenticated TaskFlow ingress for trusted external automation

Optional dashboard workboard for agent-owned cards and session handoff

Zalo Personal plugin: QR login + messaging via native zca-js (plugin install + channel config + tool)

Zoom meetings plugin: join meetings as a Chrome browser guest

Run OpenClaw embedded agent turns through the external GitHub Copilot SDK harness

OpenClaw 공식 문서 CLAUDE (CLAUDE) 한국어 가이드

How OpenClaw structures the built-in agent runtime: code layout, boundaries, resource manifests, and runtime selection.

Canonical credential eligibility and resolution semantics for auth profiles

Redirect to /tools/brave-search

Date and time handling across envelopes, prompts, tools, and connectors

Generated heading map for OpenClaw docs pages

Developer workflow for OpenClaw agent runtime: build, test, and live validation

Redirect to /tools/perplexity-search

OpenProse is a markdown-first workflow format for multi-agent AI sessions. In OpenClaw it ships as a plugin with a /prose slash command and a skill pack.

Migration plan for making ACP session and ACPX process ownership explicit

Plan and audit checklist for moving Canvas out of core and into a bundled experimental plugin.

Migration plan for making SQLite the primary durable state and cache layer while keeping config file-backed

Design for durable, deep-linkable approvals across Control UI, native apps, channels, and parent sessions

Default OpenClaw agent instructions and skills roster for the personal assistant setup

Release lanes, operator checklist, validation boxes, version naming, and cadence

Audit what can spend money, which keys are used, and how to view usage

Project origin, contributors, and license.

OpenClaw SQLite database locations, schema versions, integrity checks, and downgrade recovery

How OpenClaw vendors Apple device model identifiers for friendly names in the macOS app.

Full Release Validation stages, child workflows, release profiles, rerun handles, and evidence

Memory search providers, retrieval modes, QMD, and multimodal indexing

The @openclaw/ai npm package: reusable model transports, isolated runtimes, and host policy ports

Design for live Gateway proof of the Path 3 SQLite session/transcript flip

Prompt caching knobs, merge order, provider behavior, and tuning patterns

How Barnacle and ClawSweeper feedback helps move OpenClaw pull requests through review.

Visual summary and technical evidence for the May 2026 performance, package-size, dependency, and shrinkwrap cleanup

Rich output protocol for structured media, embeds, audio hints, and replies

RPC adapters for external CLIs (signal-cli, imsg) and gateway patterns

Secret-scanner-safe placeholder conventions for docs and examples

Deep dive: session store + transcripts, lifecycle, and (auto)compaction internals

Dev agent AGENTS.md (C-3PO)

Workspace template for BOOT.md

First-run ritual for new agents

Workspace template for AGENTS.md

Agent identity record

Dev agent identity (C-3PO)

Workspace template for SOUL.md

Dev agent soul (C-3PO)

Workspace template for TOOLS.md

Dev agent tools notes (C-3PO)

User profile record

Dev agent user profile (C-3PO)

How to run tests locally (vitest) and when to use force/coverage modes

How OpenClaw builds prompt context and reports token usage + costs

Reference: provider-specific transcript sanitization and repair rules

Full reference for CLI onboarding: every step, flag, and config field

OpenClaw 공식 문서 robots.txt (robots.txt) 한국어 가이드

How to contribute to the OpenClaw threat model

OpenClaw threat model mapped to the MITRE ATLAS framework

Machine-checked security models for OpenClaw's highest-risk paths.

How OpenClaw triages, responds to, and follows up on security incidents

How to route OpenClaw runtime HTTP and WebSocket traffic through an operator-managed filtering proxy

OpenClaw 공식 문서 sitemap.xml (sitemap.xml) 한국어 가이드

Architecture and product boundary for supervising native Codex sessions from OpenClaw.

Redirect to /tools/tts

Run OpenClaw on a Linux server or cloud VPS — provider picker, architecture, and tuning

Browser-based control UI for the Gateway (chat, activity, nodes, config)

Gateway dashboard (Control UI) access and auth

Session dashboards: architecture and implementation plan (technical design, pre-GA)

Session dashboards: agent-built widgets, boards, tabs, and the docked chat

A lobster sometimes visits the Control UI. This page explains just enough.

Terminal UI (TUI): connect to the Gateway or run locally in embedded mode

Canonical supported vs unsupported SecretRef credential surface

CLI reference for `openclaw agent` (send one agent turn via the Gateway)

CLI reference for `openclaw agents` (list/add/delete/bindings/bind/unbind/set identity)

CLI reference for `openclaw approvals` and `openclaw exec-policy`

CLI reference for metadata-only run, tool, and message lifecycle audit records

CLI reference for `openclaw browser` (lifecycle, profiles, tabs, actions, state, and debugging)

CLI reference for `openclaw clawbot` (legacy alias namespace)

Create, add, update, and remove experimental Claw agent packages

CLI reference for `openclaw completion` (generate/install shell completion scripts)

CLI reference for `openclaw configure` (interactive configuration prompts)

Redirect to the OpenClaw system-agent reference

CLI reference for `openclaw daemon` (legacy alias for gateway service management)

CLI reference for `openclaw dashboard` (open the Control UI)

CLI reference for `openclaw devices` (device pairing + token rotation/revocation)

CLI reference for `openclaw directory` (self, peers, groups)

CLI reference for `openclaw dns` (wide-area discovery helpers)

CLI reference for `openclaw docs` (search the live docs index)

CLI reference for `openclaw health` (gateway health snapshot via RPC)

CLI reference for `openclaw hooks` (agent hooks)

Expose OpenClaw channel conversations over MCP and manage saved MCP server definitions

CLI reference for `openclaw message` (send + channel actions)

CLI reference for `openclaw migrate` (import state from another agent system)

CLI reference for `openclaw node` (headless node host)

CLI reference for `openclaw nodes` (status, pairing, invoke, camera/canvas/screen/location/notify)

CLI reference and security model for the inference-backed OpenClaw setup and repair helper

CLI reference for `openclaw pairing` (approve/list pairing requests)

CLI reference for `openclaw plugins` (init, build, validate, list, install, marketplace, uninstall, enable/disable, doctor)

CLI reference for `openclaw promos` (list and claim promotional model offers)

CLI reference for `openclaw proxy`, including operator-managed proxy validation and the local debug proxy capture inspector

CLI reference for `openclaw qr` (generate mobile pairing QR + setup code)

CLI reference for `openclaw reset` (reset local state/config)

Manage sandbox runtimes and inspect effective sandbox policy

CLI reference for `openclaw secrets` (reload, audit, configure, apply)

CLI reference for `openclaw security` (audit and fix common security footguns)

CLI reference for `openclaw sessions` (list stored sessions + usage)

CLI reference for `openclaw setup` (system-agent chat with onboarding fallback)

CLI reference for `openclaw tui` (Gateway-backed or local embedded terminal UI)

CLI reference for `openclaw uninstall` (remove gateway service + local data)

CLI reference for `openclaw voicecall` (voice-call plugin command surface)

CLI reference for `openclaw webhooks` (Gmail Pub/Sub setup and runner)

Internal operator reference for the restricted cloud worker runtime

CLI reference for `openclaw commitments` (inspect and dismiss inferred follow-ups)

CLI reference for provisioning and managing isolated per-tenant OpenClaw cells

CLI reference for `openclaw path` (inspect and edit workspace files via the `oc://` addressing scheme)

CLI reference for `openclaw policy` conformance checks

CLI reference for `openclaw transcripts` (list, show, and export stored transcripts)

CLI reference for `openclaw workboard` cards, dispatch, and worker runs

Historical Node + tsx "__name is not a function" crash and its cause

Diagnostics flags for targeted debug logs

Debugging tools: watch mode, raw model streams, and tracing reasoning leakage

Frequently asked questions about OpenClaw setup, configuration, and usage

FAQ: quick-start and first-run setup — install, onboard, auth, subscriptions, initial failures

FAQ: model defaults, selection, aliases, switching, failover, and auth profiles

Repository scripts: purpose, scope, and safety notes

Testing kit: unit/e2e/live suites, Docker runners, and what each test covers

Live (network-touching) tests: model matrix, CLI backends, ACP, media providers, credentials

How OpenClaw validates update paths, package migrations, and plugin install/update behavior

OpenClaw 공식 providers/* 문서에 대응하는 한국어 가이드 인덱스입니다. 인증, 모델, 설정 키는 각 provider 페이지와 공식 문서를 확인하세요.

Alibaba Model Studio provider 설정 가이드. Alibaba Model Studio Wan video generation in OpenClaw

Anthropic provider 설정 가이드. Use Anthropic Claude via API keys or Claude CLI in OpenClaw

Arcee AI provider 설정 가이드. Arcee AI setup (auth + model selection)

Azure Speech provider 설정 가이드. Azure AI Speech text-to-speech for OpenClaw replies

Baseten provider 설정 가이드. Baseten setup for Inkling and hosted Model APIs

Amazon Bedrock provider 설정 가이드. Use Amazon Bedrock (Converse API) models with OpenClaw

Amazon Bedrock Mantle provider 설정 가이드. Use Amazon Bedrock Mantle OpenAI-compatible and Claude Messages models with OpenClaw

Cerebras provider 설정 가이드. Cerebras setup (auth + model selection)

Chutes provider 설정 가이드. Chutes setup (OAuth or API key, model discovery, aliases)

Claude Max API Proxy provider 설정 가이드. Community proxy to expose Claude subscription credentials as an OpenAI-compatible endpoint

ClawRouter provider 설정 가이드. Route credential-scoped models through ClawRouter and show managed quotas

Cloudflare AI Gateway provider 설정 가이드. Cloudflare AI Gateway setup (auth + model selection)

Cohere provider 설정 가이드. Cohere setup (auth + model selection)

ComfyUI provider 설정 가이드. ComfyUI workflow image, video, and music generation setup in OpenClaw

Deepgram provider 설정 가이드. Deepgram transcription for inbound voice notes

DeepInfra provider 설정 가이드. Use DeepInfra's unified API to access the most popular open source and frontier models in OpenClaw

DeepSeek provider 설정 가이드. DeepSeek setup (auth + model selection)

ds4 provider 설정 가이드. Run OpenClaw through ds4, a local DeepSeek V4 Flash OpenAI-compatible server

ElevenLabs provider 설정 가이드. Use ElevenLabs speech, Scribe STT, and realtime transcription with OpenClaw

fal provider 설정 가이드. fal image, video, and music generation setup in OpenClaw

Featherless AI provider 설정 가이드. Featherless AI setup, model selection, and tool calling

Fireworks provider 설정 가이드. Fireworks setup (auth + model selection)

GitHub Copilot provider 설정 가이드. Sign in to GitHub Copilot from OpenClaw using the device flow or non-interactive token import

GMI Cloud provider 설정 가이드. Use GMI Cloud's OpenAI-compatible API with OpenClaw

Google (Gemini) provider 설정 가이드. Google Gemini setup (API key + OAuth, image generation, media understanding, TTS, web search)

Gradium provider 설정 가이드. Use Gradium text-to-speech in OpenClaw

Groq provider 설정 가이드. Groq setup (auth + model selection + Whisper transcription)

Hugging Face provider 설정 가이드. Hugging Face Inference setup (auth + model selection)

inferrs provider 설정 가이드. Run OpenClaw through inferrs (OpenAI-compatible local server)

Inworld provider 설정 가이드. Inworld streaming text-to-speech for OpenClaw replies

Kilo Gateway provider 설정 가이드. Use Kilo Gateway's unified API to access many models in OpenClaw

LiteLLM provider 설정 가이드. Run OpenClaw through LiteLLM Proxy for unified model access and cost tracking

LM Studio provider 설정 가이드. Run OpenClaw with LM Studio

LongCat provider 설정 가이드. LongCat API setup for LongCat-2.0

Meta provider 설정 가이드. Meta setup (auth + muse-spark-1.1 model selection)

MiniMax provider 설정 가이드. Use MiniMax models in OpenClaw

Mistral provider 설정 가이드. Use Mistral models and Voxtral transcription with OpenClaw

Model providers provider 설정 가이드. Model providers (LLMs) supported by OpenClaw

Moonshot AI provider 설정 가이드. Configure Moonshot Kimi models vs Kimi Coding (separate providers + keys)

NovitaAI provider 설정 가이드. Use NovitaAI's OpenAI-compatible API with OpenClaw

NVIDIA provider 설정 가이드. Use NVIDIA's OpenAI-compatible API in OpenClaw

Ollama provider 설정 가이드. Run OpenClaw with Ollama (cloud and local models)

Ollama Cloud provider 설정 가이드. Use Ollama Cloud directly with OpenClaw

OpenAI provider 설정 가이드. Use OpenAI via API keys or Codex subscription in OpenClaw

OpenCode provider 설정 가이드. Use OpenCode Zen and Go catalogs with OpenClaw

OpenCode Go provider 설정 가이드. Use the OpenCode Go catalog with the shared OpenCode setup

OpenRouter provider 설정 가이드. Use OpenRouter's unified API to access many models in OpenClaw

Perplexity provider 설정 가이드. Perplexity web search provider setup (API key, search modes, filtering)

PixVerse provider 설정 가이드. PixVerse video generation setup in OpenClaw

Qianfan provider 설정 가이드. Use Qianfan's unified API to access many models in OpenClaw

Qwen provider 설정 가이드. Use Qwen Cloud through its OpenClaw plugin

Runway provider 설정 가이드. Runway video generation setup in OpenClaw

SenseAudio provider 설정 가이드. SenseAudio batch speech-to-text for inbound voice notes

SGLang provider 설정 가이드. Run OpenClaw with SGLang (OpenAI-compatible self-hosted server)

StepFun provider 설정 가이드. Use StepFun models with OpenClaw

Synthetic provider 설정 가이드. Use Synthetic's Anthropic-compatible API in OpenClaw

Tencent Cloud provider 설정 가이드. Tencent Cloud TokenHub and TokenPlan setup for hy3

Together AI provider 설정 가이드. Together AI setup (auth + model selection)

Venice AI provider 설정 가이드. Use Venice AI privacy-focused models in OpenClaw

Vercel AI Gateway provider 설정 가이드. Vercel AI Gateway setup (auth + model selection)

vLLM provider 설정 가이드. Run OpenClaw with vLLM (OpenAI-compatible local server)

Volcengine (Doubao) provider 설정 가이드. Volcano Engine setup (Doubao models, coding endpoints, and Seed Speech TTS)

Vydra provider 설정 가이드. Use Vydra image, video, and speech in OpenClaw

xAI provider 설정 가이드. Use xAI Grok models in OpenClaw

Xiaomi MiMo provider 설정 가이드. Use Xiaomi MiMo pay-as-you-go and Token Plan models with OpenClaw

Z.AI provider 설정 가이드. Use Z.AI (GLM models) with OpenClaw

OpenClaw 공식 tools/* 문서에 대응하는 한국어 가이드 인덱스입니다. browser, exec, search, media, agent 조율 도구를 포함합니다.

ACP agents 도구 가이드. Run external coding harnesses (Claude Code, Cursor, Gemini CLI, explicit Codex ACP, OpenClaw ACP, OpenCode) through the ACP backend

ACP agents — setup 도구 가이드. Setting up ACP agents: acpx harness config, plugin setup, permissions

Agent send 도구 가이드. Run agent turns from the CLI and optionally deliver replies to channels

apply_patch tool 도구 가이드. Apply multi-file patches with the apply_patch tool

Ask user 도구 가이드. How ask_user pauses an agent turn for a structured human decision

Brave search 도구 가이드. Brave Search API setup for web_search

Browser (OpenClaw-managed) 도구 가이드. Integrated browser control service + action commands

Browser control API 도구 가이드. OpenClaw browser control API, CLI reference, and scripting actions

Browser troubleshooting 도구 가이드. Fix Chrome/Brave/Edge/Chromium CDP startup issues for OpenClaw browser control on Linux

Browser login 도구 가이드. Manual logins for browser automation + X/Twitter posting

WSL2 + Windows + remote Chrome CDP troubleshooting 도구 가이드. Troubleshoot WSL2 Gateway + Windows Chrome remote CDP in layers

BTW side questions 도구 가이드. Ephemeral side questions with /btw

Capabilities (이동됨) 도구 가이드. Redirect to /plugins/adding-capabilities

Chrome Extension 도구 가이드. Chrome extension: let OpenClaw drive your signed-in Chrome with no remote-debugging prompt

ClawHub (이동됨) 도구 가이드. Redirect to /clawhub

Code execution 도구 가이드. code_execution: run sandboxed remote Python analysis with xAI

Code Mode 도구 가이드. Use OpenClaw Code Mode to discover, call, and combine large tool catalogs in compact JavaScript or TypeScript workflows

Creating skills 도구 가이드. Build, test, and publish custom SKILL.md workspace skills for your OpenClaw agents.

Diffs 도구 가이드. Read-only diff viewer and file renderer for agents (optional plugin tool)

DuckDuckGo search 도구 가이드. DuckDuckGo web search -- key-free provider (experimental, HTML-based)

Elevated mode 도구 가이드. Elevated exec mode: run commands outside the sandbox from a sandboxed agent

Exa search 도구 가이드. Exa AI search -- neural and keyword search with content extraction

Exec tool 도구 가이드. Exec tool usage, stdin modes, and TTY support

Exec approvals 도구 가이드. Host exec approvals: policy knobs, allowlists, and the YOLO/strict workflow

Exec approvals — advanced 도구 가이드. Advanced exec approvals: safe bins, interpreter binding, approval forwarding, native delivery

Firecrawl 도구 가이드. Firecrawl search, scrape, and web_fetch fallback

Gemini search 도구 가이드. Gemini web search with Google Search grounding

Goal 도구 가이드. Session goals: durable per-session objectives, /goal controls, model goal tools, token budgets, and TUI status

Grok search 도구 가이드. Grok web search via xAI web-grounded responses

Image generation 도구 가이드. Generate and edit images via image_generate across OpenAI, Google, fal, Microsoft Foundry, MiniMax, ComfyUI, DeepInfra, OpenRouter, LiteLLM, xAI, Vydra

Kimi search 도구 가이드. Kimi web search via Moonshot web search

LLM task 도구 가이드. JSON-only LLM tasks for workflows (optional plugin tool)

Lobster 도구 가이드. Typed workflow runtime for OpenClaw with resumable approval gates.

Tool-loop detection 도구 가이드. How to enable guardrails that detect repetitive tool-call loops

Media overview 도구 가이드. Image, video, music, speech, and media-understanding capabilities at a glance

MiniMax search 도구 가이드. MiniMax Search via the Token Plan search API

Multi-agent sandbox and tools 도구 가이드. Per-agent sandbox + tool restrictions, precedence, and examples

Music generation 도구 가이드. Generate music via music_generate across ComfyUI, fal, Google Lyria, MiniMax, and OpenRouter workflows

Ollama web search 도구 가이드. Ollama Web Search via a local Ollama host or the hosted Ollama API

Parallel search 도구 가이드. Parallel Search -- LLM-optimized dense excerpts from web sources

PDF tool 도구 가이드. Analyze one or more PDF documents with native provider support and extraction fallback

Permission modes 도구 가이드. Permission modes for host exec, Codex Guardian approvals, and ACPX harness sessions

Perplexity search 도구 가이드. Perplexity Search API and Sonar/OpenRouter compatibility for web_search

Plugins 도구 가이드. Install, configure, and manage OpenClaw plugins

Reactions 도구 가이드. Reaction tool semantics across all supported channels

Screen 도구 가이드. Let an agent arrange the connected Control UI

SearXNG search 도구 가이드. SearXNG web search -- self-hosted, key-free meta-search provider

Self-learning 도구 가이드. Let OpenClaw propose reusable skills from corrections and substantial completed work

Show widget 도구 가이드. Show self-contained HTML widgets on supported chat surfaces

Skill Workshop 도구 가이드. Create and update workspace skills through Skill Workshop review

Slash commands 도구 가이드. All available slash commands, directives, and inline shortcuts — configuration, routing, and per-surface behavior.

Steer 도구 가이드. Steer an active run without changing queue mode

Swarm 도구 가이드. Orchestrate concurrent sub-agents from Code Mode scripts with structured results, bounded fan-out, and live progress

Tavily 도구 가이드. Tavily search and extract tools

Thinking levels 도구 가이드. Directive syntax for /think, /fast, /verbose, /trace, and reasoning visibility

Tokenjuice 도구 가이드. Compact noisy exec and bash tool results with the optional Tokenjuice plugin

Tool Search 도구 가이드. Tool Search: compact large OpenClaw tool catalogs behind search, describe, and call

Trajectory bundles 도구 가이드. Export redacted trajectory bundles for debugging an OpenClaw agent session

Text-to-speech 도구 가이드. Text-to-speech for outbound replies — providers, personas, slash commands, and per-channel output

Video generation 도구 가이드. Generate videos via video_generate from text, image, or video references across 16 provider backends

Web search 도구 가이드. web_search, x_search, and web_fetch -- search the web, search X posts, or fetch page content

Web fetch 도구 가이드. web_fetch tool -- HTTP fetch with readable content extraction