Tools
Manifest Self
By Smart LLM Routing for OpenClaw. Cut Costs up to 70% 🦞🦚 (fork — manifest-self plugin)
Install
npm install &&README
<p align="center">
<picture>
<source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/mnfst/manifest/HEAD/.github/assets/logo-white.svg" />
<source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/mnfst/manifest/HEAD/.github/assets/logo-dark.svg" />
<img src="https://raw.githubusercontent.com/mnfst/manifest/HEAD/.github/assets/logo-dark.svg" alt="Manifest" height="53" title="Manifest"/>
</picture>
</p>
<p align="center">
Manifest Self — OpenClaw Plugin Fork
</p>
> **This is a fork of [mnfst/manifest](https://github.com/mnfst/manifest)** focused on local OpenClaw plugin development. For Docker self-hosting or the cloud version, use the [upstream repo](https://github.com/mnfst/manifest) directly.
## What is this fork?
[Manifest](https://manifest.build) is a smart model router for AI agents — it scores each request and routes it to the cheapest model that can handle it, saving up to 70% on LLM costs.
Upstream discontinued the OpenClaw plugin approach ([#1527](https://github.com/mnfst/manifest/pull/1527), [#1528](https://github.com/mnfst/manifest/pull/1528), [#1539](https://github.com/mnfst/manifest/pull/1539)) because embedding a full NestJS server inside the gateway process caused lifecycle conflicts. **This fork fixes that problem** with `manifest-self` — a lightweight plugin that runs the backend as a separate process instead of embedding it.
| | Old embedded plugin (discontinued) | `manifest-self` (this fork) |
|---|---|---|
| Architecture | NestJS inside gateway process | Separate detached process |
| Plugin size | ~tens of MB (30+ deps) | 14.8 KB (zero deps) |
| Lifecycle issues | Double-boot, orphan processes, SQLite locking | Clean process boundary |
| Status | Deprecated by upstream | Actively developed |
## Quick start
```bash
# Clone and build
git clone https://github.com/andychu666/manifest-self.git
cd manifest-self && git checkout openclaw-plugin-v5.45.1
npm install && npm run build
# Install in OpenClaw
openclaw plugins install --path ./packages/openclaw-plugins/manifest-self
openclaw config set plugins.entries.manifest-self.enabled true
openclaw gateway restart
```
The plugin auto-starts the backend on port 2099. Open the dashboard at [http://localhost:2099](http://localhost:2099) to connect your LLM providers and configure routing.
Full documentation: [`packages/openclaw-plugins/manifest-self/README.md`](packages/openclaw-plugins/manifest-self/README.md)
## How it works
Every request to `manifest/auto` goes through a 23-dimension scoring algorithm (runs in under 2ms). The scorer picks a tier (simple, standard, complex, or reasoning) and routes to the best model in that tier from your connected providers.
All routing data (tokens, costs, model, latency) is recorded automatically. You see it in the dashboard. No extra setup.
## Manifest vs OpenRouter
| | Manifest | OpenRouter |
| ------------ | -------------------------------------------- | --------------------------------------------------- |
| Built for | Personal AI agents and consumer apps | Enterprise API traffic |
| Architecture | Local. Your requests, your providers | Cloud proxy. All traffic goes through their servers |
| Cost | Free | 5% fee on every API call |
| Source code | MIT, fully open | Proprietary |
| Data privacy | Metadata only (cloud) or fully local | Prompts and responses pass through a third party |
| Transparency | Open scoring. You see why a model was chosen | No visibility into routing decisions |
| Control | Define your model per tier with up to 5 fallbacks each, from complexity tiers (Simple → Reasoning) to specialized ones (Coding, Vision). | Flat fallback list per request, or opaque auto-routing. No user-defined tiers. |
| Custom providers | Add custom providers and models | Supported providers only, no arbitrary endpoints |
| Subscription support | Route through flat-rate subscriptions you already pay for (MiniMax $20/mo, etc.) | Pay-per-use billing |
## Supported providers
Works with 300+ models. Connect with an API key or reuse a paid subscription:
| Provider | API key | Subscription |
|----------|:-------:|:-------------|
| [OpenAI](https://platform.openai.com/) | ✅ | ✅ ChatGPT Plus / Pro / Team |
| [Anthropic](https://www.anthropic.com/) | ✅ | ✅ Claude Max / Pro |
| [Google Gemini](https://ai.google.dev/) | ✅ | |
| [DeepSeek](https://www.deepseek.com/) | ✅ | |
| [xAI](https://x.ai/) | ✅ | |
| [Mistral AI](https://mistral.ai/) | ✅ | |
| [OpenRouter](https://openrouter.ai/) | ✅ | |
| [LM Studio](https://lmstudio.ai/) | ✅ Local | |
| [Ollama](https://ollama.com/) | ✅ Local | ✅ Ollama Cloud |
| Custom providers (OpenAI-compatible) | ✅ | |
## Other ways to use Manifest
This fork only maintains the **local OpenClaw plugin**. For other deployment options, use upstream:
| Method | Where |
|--------|-------|
| **Cloud** (hosted) | [app.manifest.build](https://app.manifest.build) |
| **Docker** (self-hosted) | [upstream repo](https://github.com/mnfst/manifest) / [Docker Hub](https://hub.docker.com/r/manifestdotbuild/manifest) |
## Development
```bash
# Watch mode — rebuilds plugin on source changes
cd packages/openclaw-plugins/manifest-self && npm run dev
# After changes, restart gateway
openclaw gateway restart
```
## License
[MIT](LICENSE)
tools
Comments
Sign in to leave a comment