Documentation Index
Fetch the complete documentation index at: /llms.txt
Use this file to discover all available pages before exploring further.

Skip to main content

OpenClaw home page

GitHub
Releases
Discord

Get started

CLI commands

CLI reference

RPC and API

RPC adapters
External apps
Device model database

Codex harness

Codex harness reference
Codex harness runtime

Plugin reference

Plugin inventory
Plugin reference
Dependencies
Install overrides

Plugin SDK reference

Plugin SDK overview
Plugin SDK subpaths
Entry Points
Runtime helpers
Agent Harness
Setup and config
Testing
Plugin manifest

Plugin maintainer reference

Internals
Plugin architecture internals
Migrate to SDK
Plugin compatibility
Channel outbound API
Channel inbound API
Channel Ingress
Message presentation

Templates

Default AGENTS.md
AGENTS.md template
BOOT.md template
BOOTSTRAP.md template
Retired HEARTBEAT.md workspace file
IDENTITY template
SOUL.md template
USER template

Technical reference

Agent runtime architecture
Token use and costs
SecretRef credential surface
@openclaw/ai package
Prompt caching
API usage and costs
Transcript hygiene
Database schemas
Memory config
Rich output protocol
Session management deep dive
Date and time

Concept internals

TypeBox
Markdown formatting
Typing indicators
Usage tracking
Timezones

Project

Credits

Contributing

PR review flow

vLLM plugin
Distribution
Surface
Related docs

vLLM plugin

Adds vLLM model provider support to OpenClaw.

Distribution

Package: @openclaw/vllm-provider
Install route: included in OpenClaw

Surface

providers: vllm

Related docs

vllm

Suggest edits

Assistant

Responses are generated using AI and may contain mistakes.

​vLLM plugin

​Distribution

​Surface

​Related docs

vLLM plugin

Distribution

Surface

Related docs