OpenClaw Consensus
9-LLM consensus + disagreement scoring + cheapest-route picks to fight hallucinations.
Open source Open in the app JSON README (API)
About
9-LLM consensus + disagreement scoring + cheapest-route picks to fight hallucinations.
Details
- Kind
- MCP servers
- Topic
- AI, RAG & memory
- Publisher
- miconnm
- Origin
- official
- Category
- ferramentas
- Transport
- local
- Version
- 0.1.1
- Open pull requests
- 4
- Last push
- 2026-08-01T08:46:09Z
- Repository state
- ativo
- Language
- Python
- License
- MIT
- Added
- 2026-08-29 03:02:05
- Updated
- 2026-08-29 03:02:05
- Origin id
io.github.MICONNM/openclaw-consensus-mcp
README
# OpenClaw Consensus MCP
[](https://github.com/MICONNM/openclaw-consensus-mcp/actions/workflows/ci.yml)
[](https://pypi.org/project/openclaw-consensus-mcp/)
[](LICENSE)
> Multi-model consensus inside MCP clients: compare answers, surface disagreement, and escalate only when needed.
OpenClaw Consensus MCP wraps the OpenClaw Consensus API as three Model Context Protocol tools. It is designed for workflows where a maintainer wants a second opinion before accepting a risky answer, review summary, or routing decision.
<!-- mcp-name: io.github.MICONNM/openclaw-consensus-mcp -->
## What it does
OpenClaw runs the same prompt across multiple models, then returns:
- a **consensus answer** with confidence and model response metadata,
- a **disagreement heuristic** derived from the deep consensus response, and
- a **cheapest route** recommendation that tries smaller model sets before escalating.
This MCP server exposes those three capabilities as tools so Claude Desktop / Claude Code can call them mid-conversation.
## Why consensus?
A single model can produce a confident but incorrect answer. Comparing multiple responses does not prove correctness, but disagreement is a useful signal that a maintainer should review the output more carefully.
## Install
```bash
pip install openclaw-consensus-mcp
# or
uv pip install openclaw-consensus-mcp
```
You also need a RapidAPI key for the OpenClaw Consensus API:
<https://rapidapi.com/yanmiayn/api/openclaw-consensus>
Set it in your environment:
```bash
export RAPIDAPI_KEY="your-rapidapi-key"
```
## Claude Desktop config
Add to `~/.claude/claude_desktop_config.json` (macOS/Linux) or
`%APPDATA%\Claude\claude_desktop_config.json` (Windows):
```json
{
"mcpServers": {
"openclaw-consensus": {
"command": "openclaw-consensus",
"env": {
"RAPIDAPI_KEY": "your-rapidapi-key"
}
}
}
}
```
For Claude Code:
```bash
claude mcp add openclaw-consensus -- openclaw-consensus
```
## Tools
### `consensus(prompt, mode="balanced")`
Get a 9-LLM consensus answer.
- **prompt** *(string)* — the question.
- **mode** *(string, default `balanced`)* — `deep` (9 models), `balanced` (5), or `fast` (3).
**Returns**
```json
{
"consensus": "string",
"confidence": 0.0,
"models_responded": 5,
"votes": []
}
```
The `consensus` tool returns the upstream API response as-is. Fields may expand as the endpoint evolves.
### `disagreement_score(prompt)`
How much the deep consensus response disagrees on a prompt.
**Returns**
```json
{
"disagreement": 0.0,
"confidence": 1.0,
"models_responded": 9,
"votes": []
}
```
### `cheapest_route(prompt, target_quality=0.85)`
Try `fast`, `balanced`, and `deep` modes in order until the confidence threshold is met.
**Returns**
```json
{
"selected_mode": "balanced",
"models_used": 5,
"confidence": 0.9,
"answer": "string"
}
```
## Local development
```bash
git clone https://github.com/MICONNM/openclaw-consensus-mcp
cd openclaw-consensus-mcp
uv venv && source .venv/bin/activate
uv pip install -e ".[dev]"
pytest
```
Smoke-test the server with the official MCP Inspector:
```bash
npx @modelcontextprotocol/inspector openclaw-consensus
```
## Publish
```bash
uv build
uv publish # to PyPI
mcp-publisher publish # to the official MCP Registry
```
See [CONTRIBUTING.md](CONTRIBUTING.md) for the development workflow and [docs/maintainer-workflow.md](docs/maintainer-workflow.md) for triage, review, security, and release responsibilities.
## Limitations
- Consensus is a review aid, not a correctness guarantee.
- Network-backed tools require a configured OpenClaw endpoint and may incur provider charges.
- Do not send secrets, private source code, or personal data unless your endpoint policy explicitly allows it.
## Security
Please report vulnerabilities privately using the process in [SECURITY.md](SECURITY.md).
## License
MIT — see [LICENSE](LICENSE).