OpenClaw.NET

中文

Disclaimer: This project is not affiliated with, endorsed by, or associated with OpenClaw. It is an independent .NET implementation inspired by their work.

OpenClaw.NET is a NativeAOT-friendly AI agent runtime and gateway for .NET with practical OpenClaw ecosystem compatibility.

It is for .NET developers and operators who want a local or self-hosted agent gateway with explicit diagnostics, first-party .NET tools, OpenAI-compatible HTTP surfaces, and a path from source checkout to NativeAOT release artifacts.

Docs: AgentQi.dev is the documentation and ecosystem home for OpenClaw.NET. OpenClaw.NET remains the current runtime and repository identity.

AgentQi Documentation

AgentQi is the broader developer-infrastructure direction behind OpenClaw.NET: practical, observable, self-hosted AI agent systems for .NET developers.

OpenClaw.NET is the runtime and repository you can use today. AgentQiX is the likely future runtime identity.

Start here:

• Quickstart
• Getting Started
• Architecture
• Security
• Roadmap

What Works Now

• NativeAOT-friendly runtime and gateway for .NET agent workloads
• Agent runtime with tool execution, streaming, cancellation, retry, memory, and session support
• Gateway with chat UI, admin UI, OpenAI-compatible endpoints, MCP, websocket, health, and diagnostics
• Passive Harness Contracts for inspectable agent-work plans without changing default chat or approval behavior
• Passive Evidence Bundles for inspectable run evidence, checks, risks, and human review without default runtime interception
• Optional Plan-Execute-Verify Mode for governed high-risk tool execution with contracts, evidence, and verification
• Passive Governance Ledger for durable approval and oversight decisions without auto-approving future actions
• Harness Regression Suite via openclaw harness test for offline checks before trusting harness/runtime changes
• Harness Evolution Proposals for review-first suggestions to improve harness policies, routing, memory retrieval, verification, pulse behavior, and tool governance
• Optional Fractal Memory MCP integration for compact structured project memory and Runtime Pulse context without replacing OpenClaw memory/session stores
• Shared Harness State for passive delegated-work coordination across participants, actions, read/write sets, assumptions, verifier obligations, evidence links, and conflicts
• Codebase Harness Map via openclaw harness map for passive static repository maps of projects, modules, endpoints, tools, providers, channels, config, and tests
• OpenClaw SkillKit via openclaw skill for local-first skill authoring, validation, critique, packaging, and dry-run execution planning
• First-class MCP App support for manifest-discovered third-party MCP apps with lifecycle management, tool bridging, and interactive UI resources
• Session-scoped /goal auto-continuation for long-running work that should continue until complete, blocked, or budget-limited
• TokenJuice output reduction for deterministic, rule-driven compression of verbose tool output before it enters model context
• First-class optional Microsoft Agent Framework adapter for Runtime.Orchestrator=maf without a special build
• Durable workflow delegation through supported workflow backends such as maf-durable-http
• CLI and Companion setup flows for source checkouts and desktop bundles
• /loop recurring-prompt command with TickerQ-backed session-scoped timer injection, idempotent override, and dual-path semantic auto-termination for build health checks, log polling, and other periodic tasks
• 80+ native and optional tool surfaces covering file ops, sessions, memory, web, messaging, home automation, databases, email, MCP apps, and more
• 9 channel adapters (Telegram, SMS, WhatsApp, Teams, Slack, Discord, Signal, email, webhooks) with DM policy, allowlists, and signature validation
• Native LLM providers for OpenAI, Claude, Gemini, Azure OpenAI, Ollama, and OpenAI-compatible endpoints
• Optional embedded local models with Gemma 4 GGUF packages, package install/verify CLI commands, supervised sidecar inference, and frame-based video understanding
• Practical reuse of existing OpenClaw TS/JS plugins and SKILL.md packages

Start with docs/START_HERE.md for the evaluator overview, docs/QUICKSTART.md for the supported local setup path, or docs/RELEASES.md for desktop downloads.

For Microsoft Agent Framework, A2A, and durable workflow setup, see docs/integrations/microsoft-agent-framework.md, docs/a2a.md, and docs/workflow-backends.md.

Download And Run Desktop

For the lowest-friction desktop start, download the latest desktop bundle for your platform:

Each desktop bundle includes Companion, the NativeAOT gateway, and the NativeAOT CLI.

1. Extract the archive.
2. Launch Companion from the companion folder.
3. Open the Setup tab.
4. Choose a provider/model and enter the provider key, choose Ollama for a local model server, or choose Embedded for an OpenClaw-managed local model such as Gemma 4.
5. Click Set Up and Start.

Companion writes a local config, starts the bundled gateway on 127.0.0.1, and connects to it. The current Windows and macOS release archives are unsigned, so first-run OS warnings are expected. See docs/RELEASES.md for checksums, standalone CLI/gateway archives, signing status, and maintainer release flow.

Quickstart

For a real local gateway from source:

export MODEL_PROVIDER_KEY="sk-..."
dotnet run --project src/OpenClaw.Cli -c Release -- start

When the gateway finishes startup it now prints explicit phase markers, a final OpenClaw gateway ready. block, the localhost URLs, Ctrl-C to stop, and any non-fatal startup notices under Started with notices:. Then open:

The root URL redirects to /chat. For the full first-run walkthrough (including the "First 10 Minutes" runbook and debugging flow), see docs/QUICKSTART.md. For the project shape and repository map before changing code, see docs/GETTING_STARTED.md.

If you want a direct gateway fallback instead of the full CLI onboarding flow, run:

dotnet run --project src/OpenClaw.Gateway -c Release -- --quickstart

--quickstart is interactive-only. It applies a minimal loopback-local profile for the current process, prompts for missing provider inputs, retries on the common first-run failures, and after a successful start can save the working setup to ~/.openclaw/config/openclaw.settings.json.

If the CLI is already on your PATH, the same guided entrypoints are:

openclaw start
openclaw setup
openclaw setup launch --config ~/.openclaw/config/openclaw.settings.json
openclaw setup service --config ~/.openclaw/config/openclaw.settings.json --platform all
openclaw setup status --config ~/.openclaw/config/openclaw.settings.json
openclaw upgrade check --config ~/.openclaw/config/openclaw.settings.json
openclaw upgrade rollback --config ~/.openclaw/config/openclaw.settings.json --offline

Useful follow-up commands and surfaces:

openclaw models presets
openclaw models packages
openclaw models install gemma-4-e4b --accept-license --path ~/Downloads/gemma-4-E4B-it-Q4_K_M.gguf --mmproj-path ~/Downloads/mmproj-gemma-4-E4B-it-Q8_0.gguf
openclaw models doctor
openclaw maintenance scan --config ~/.openclaw/config/openclaw.settings.json
openclaw maintenance fix --config ~/.openclaw/config/openclaw.settings.json --dry-run
openclaw skill new "Community Research Insight Extractor" --category research
openclaw skill validate community.research_insight
openclaw skills inspect ./skills/my-skill
openclaw compatibility catalog
openclaw insights
openclaw admin trajectory export --anonymize --output ./trajectory.jsonl
openclaw upgrade check --config ~/.openclaw/config/openclaw.settings.json --offline
openclaw upgrade rollback --config ~/.openclaw/config/openclaw.settings.json --offline
openclaw migrate upstream --source ./upstream-agent --target-config ~/.openclaw/config/openclaw.settings.json

• Sk