OpenClaw.NET Documentation Index
June 29, 2026 · View on GitHub
Use this page as the map. If you are unsure where to go next, the groups below are roughly the order most people need them.
Start Here
| Doc | When to read |
|---|---|
| START_HERE.md | You are evaluating the repo and want the shortest orientation: what works, what is experimental, how to prove the runtime, and where the compatibility boundaries are. |
| SITE_MAP.md | You are turning these Markdown files into a documentation website and need a stable navigation order. |
| GETTING_STARTED.md | You cloned the repo and want the project shape, repository map, and first-run debugging flow before running commands. |
| QUICKSTART.md | You want the shortest supported path to a running local instance. |
| GLOSSARY.md | A term in another doc is unfamiliar — gateway, runtime, skill, plugin, profile, posture, aot / jit / auto, etc. |
| ../README.md | High-level overview, feature list, and headline capabilities. |
| ../README-cn.md | 项目概述、功能列表和核心能力（简体中文）。 |
| ARCHITECTURE_BOUNDARIES.md | Core, gateway, extension, AOT/JIT, and Industrial Pack boundaries. |
| CAPABILITY_MATRIX.md | Core, optional, experimental, and JIT-only capability lanes. |
| zh-CN/CAPABILITY_MATRIX.md | 能力矩阵（简体中文）。 |
| zh-CN/START_HERE.md | Simplified Chinese first-run orientation. |
Using It
| Doc | What it covers |
|---|---|
| USER_GUIDE.md | Providers, tools, skills, channels, and the day-to-day operator surface. |
| RELEASES.md | Desktop download bundles, release assets, and signing/notarization status. |
| TOOLS_GUIDE.md | Native and optional tool catalog, behavior, and configuration. |
| MCPAPP.md | MCP App manifest discovery, lifecycle management, tool bridging, resources, prompts, and UI bundles. |
| tokenjuice.md | Deterministic, rule-driven tool output reduction before tool results enter model context. |
| GOAL_TECHNICAL_ARCHITECTURE.md | Session-scoped /goal auto-continuation architecture, states, tools, budget limits, and runtime integration. |
| LOOP_TECHNICAL_ARCHITECTURE.md | /loop recurring-prompt command architecture, scheduler behavior, control tool, and termination model. |
| zh-CN/TOOLS_GUIDE.md | 原生和可选工具目录与配置（简体中文）。 |
| zh-CN/MCPAPP.md | MCP App 支持（简体中文）。 |
| zh-CN/SITE_MAP.md | 简体中文文档地图。 |
| zh-CN/LOOP_TECHNICAL_ARCHITECTURE.md | /loop 定时循环命令技术架构（简体中文）。 |
| zh-CN/GOAL_TECHNICAL_ARCHITECTURE.md | Goal 目标机制技术架构（简体中文）。 |
| SKILLKIT.md | Local-first openclaw skill workflows for defining, validating, critiquing, packaging, and dry-running reusable OpenClaw skills. |
| LOCAL_MODELS.md | Embedded local model packages, supervised sidecars, frame-based video support, and experimental LiteRT-LM adapter guidance. |
| opensquilla-dynamic-turn-routing.md | Independent guide to the OpenSquilla-style dynamic turn-routing surface, current implementation status, runtime wiring, and operator expectations. |
| opensquilla-meta-skill-migration.md | OpenClaw.NET meta-skill migration notes for OpenSquilla-style DAG orchestration, current parity, and remaining migration gaps. |
| dynamic-turn-routing-model-profiles.md | Collaboration playbook for how Dynamic Turn Routing and Model Profiles work together (precedence, field projection, fallback, and restore semantics). |
| EXTERNAL_CLI_CONNECTORS.md | Governed external CLI connectors, optional presets, named command allowlists, approvals, redaction, and audit behavior. |
| plugins/payment.md | Native payment tool, virtual cards, machine payments, providers, and safe agent-facing actions. |
| cli/payment.md | openclaw payment ... gateway-backed CLI commands and safe output contract. |
| cli/routing.md | openclaw routing ... commands for onboarding, router modes, tier/provider mapping, and diagnostics. |
| mempalace-memory.md | Optional ElBruno.MempalaceNet memory provider and temporal KG tool. |
| FRACTAL_MEMORY.md | Optional MCP-first Fractal Memory integration for compact structured project memory and Runtime Pulse context. |
| providers/microsoft-extensions-ai.md | Optional JIT bridge for arbitrary Microsoft.Extensions.AI.IChatClient providers. |
| SESSIONS.md | Session lifecycle, the SessionManager, sessions_spawn / sessions_yield / sessions tools, and per-turn token accounting semantics (turn/session/runtime/provider views). |
| CANVAS_A2UI.md | Supported Canvas and A2UI behavior for agent-rendered visual workspaces. |
| integrations/microsoft-agent-framework.md | Supported optional Microsoft Agent Framework runtime adapter, runtime selection, A2A setup, and migration from old experimental config. |
| workflow-backends.md | Durable workflow delegation, maf-durable-http, integration API and MCP tools, status model, and sample host. |
| a2a.md | A2A v1 discovery, endpoint, authentication, and deployment contract through the Microsoft Agent Framework adapter. |
| MODEL_PROFILES.md | Provider-agnostic named model profiles, including Gemma-family setups. |
| PROMPT_CACHING.md | Provider-aware prompt caching hints, dialects, diagnostics. |
| PULSE.md | Runtime Pulse scheduled heartbeat turns, HEARTBEAT.md, alert suppression, and operator controls. |
| LEARNING.md | Review-first learning proposals, automation suggestion quality gates, feedback events, and harness regression relationships. |
| HARNESS_CONTRACTS.md | Passive, inspectable Harness Contract records for future plan-execute-verify and evidence workflows. |
| EVIDENCE_BUNDLES.md | Passive evidence records for what happened, what was checked, remaining uncertainty, and operator trust. |
| PLAN_EXECUTE_VERIFY.md | Optional governed Plan-Execute-Verify mode for high-risk tool execution. |
| GOVERNANCE_LEDGER.md | Passive approval and oversight decision history as durable harness state. |
| HARNESS_REGRESSION.md | Offline-first CLI regression checks for harness safety, onboarding, memory, provider-shape, MCP, OpenAI-compatible, and serialization guarantees. |
| HARNESS_EVOLUTION.md | Review-first proposals for harness changes to policies, routing, memory retrieval, verification, pulse behavior, and tool governance. |
| SHARED_HARNESS_STATE.md | Passive shared state for delegated and future multi-agent workflows, including participants, actions, read/write sets, assumptions, verifier obligations, and conflicts. |
| CODEBASE_HARNESS_MAP.md | Passive static repository maps for agent harness environments, including projects, modules, endpoints, tools, providers, channels, config surfaces, tests, and diagnostics. |
| governance/sidecar-pattern.md | Optional central tool-governance middleware, sidecar flow, decisions, and audit fields. |
| governance/microsoft-agent-governance.md | Microsoft Agent Governance sidecar integration notes and deployment cautions. |
| deployment/TAILSCALE.md | Optional Tailscale Serve private runtime access guidance. |
Testing and Evaluation
| Doc | What it covers |
|---|---|
| testing/agent-testing-harness.md | Scenario-based agent tests, trace artifacts, explicit oracles, CLI usage, xUnit usage, and future runtime/gateway adapter seams. |
| testing/ai-assisted-testing-playbook.md | Disciplined AI-assisted testing workflow: scenario matrices, oracle requirements, boundary cases, human review, and trace-to-regression loops. |
| testing/routing-evaluation.md | Offline dynamic routing baseline comparison, sample schema, and regression gate for alwaysT2, ruleOnly, and classifierPlusRules. |
| HARNESS_REGRESSION.md | openclaw harness test for offline regression checks before trusting harness/runtime changes. |
| MODEL_PROFILES.md#evaluation-harness | Existing gateway-backed model/profile evaluation surface exposed by openclaw eval. |
Channels and Integrations
| Doc | What it covers |
|---|---|
| TEAMS_SETUP.md | Microsoft Teams channel setup. |
| WHATSAPP_SETUP.md | First-party WhatsApp integration (Baileys and whatsmeow workers). |
| SEMANTIC_KERNEL.md | Semantic Kernel interop surface. |
| external-coding-backends.md | External coding backend integration. |
| external-coding-backends-summary.md | Summary of supported OSS external coding backends. |
Operating It
| Doc | What it covers |
|---|---|
| ../SECURITY.md | Security posture, required settings before a public deployment, and breaking-change credential rules. |
| security/payments.md | Payment approvals, vaulting, redaction, sentinel substitution, and provider security invariants. |
| COMPATIBILITY.md | Supported upstream skill, plugin, and channel surface. |
| sandboxing.md | Optional sandbox execution backends. |
| DOCKERHUB.md | Official container image reference. |
| build/dashboard-assets.md | Gateway Dashboard asset build and publish behavior. |
| deployment/TAILSCALE.md | Tailscale Serve/Funnel patterns and private runtime access security notes. |
| PRODUCTION_FIXES.md | Known production-readiness fixes and their verification. |
Extending It
| Doc | What it covers |
|---|---|
| ../CONTRIBUTING.md | Build, test, and PR expectations for contributors. |
| project/governance.md | Project governance principles, role boundaries, commercial-use disclosure, vendor neutrality, and Industrial Pack scope. |
| project/maintainers.md | Maintainer role definitions, area maintainer scope, review expectations, and approval boundaries. |
| project/sponsors.md | Sponsorship support, limits, commercial-use guidance, and sponsor listing policy. |
| project/branch-protection.md | Recommended main branch protection, required checks, and CODEOWNERS relationship. |
| maintainers/review-checklist.md | Maintainer checklist for runtime, gateway, extension, industrial, documentation, security, AOT, and commercial review. |
| ai-contributor-guide.md | Guidance for AI-assisted contribution workflows. |
| architecture-startup-refactor.md | Current gateway startup layout and composition seams. |
| architecture/optional-dependency-split.md | Optional dependency split guidance and follow-up seams. |
| proposals/industrial-pack-preview.md | Proposal for a reusable, vendor-neutral Industrial Pack preview. |
| ROADMAP.md | Planned direction and priorities. |
Research and Experiments
The experiments/ directory holds time-stamped findings from specific investigations. These are frozen snapshots, not living documentation — read them when you are tracing the history of a decision (most are about the Microsoft Agent Framework AOT/JIT work).
Doc Conventions
- Commands assume a source checkout. Replace
openclaw ...withdotnet run --project src/OpenClaw.Cli -c Release -- ...when running from source. - The supported config path is an external JSON file generated by
openclaw setup, typically at~/.openclaw/config/openclaw.settings.json. - "Posture" refers to the combined security/deployment checks surfaced by
--doctor,openclaw admin posture, andopenclaw setup status. When a doc says "check your posture", run those three.
Missing Something?
If a doc does not answer the question you came with, that is a bug. Open a documentation issue with the exact question and the doc you read, or link to this page in a PR that adds the missing piece. See the "Getting help" section at the end of GETTING_STARTED.md for the full loop.