@peacethinking/mcp-server
v0.1.0
Published
MCP server for AI-assisted conflict resolution — REST + embedded modes, 13 tools
Maintainers
Readme
@peacethinking/mcp-server
MCP server for AI-assisted conflict resolution. Connects Claude, ChatGPT, or any MCP-compatible AI agent to the Peacethinking resolution platform. 15 tools — from escalation detection to full agent-to-agent mediation.
Quick Start
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"peacethinking": {
"command": "npx",
"args": ["@peacethinking/mcp-server", "--stdio"],
"env": {
"PEACETHINKING_API_KEY": "pt_live_YOUR_KEY"
}
}
}
}Programmatic Usage
import { createPeacethinkingServer } from '@peacethinking/mcp-server';
import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js';
const server = createPeacethinkingServer({
mode: 'rest',
apiKey: 'pt_live_YOUR_KEY',
});
const transport = new StdioServerTransport();
await server.connect(transport);Installation
npm install @peacethinking/mcp-server
# or
pnpm add @peacethinking/mcp-serverOr run directly:
npx @peacethinking/mcp-server --stdioServer Modes
| Mode | Description | Use Case |
|------|-------------|----------|
| rest | HTTP calls to peacethinking.ai API | Default — CLI, Claude Desktop, external agents |
| embedded | Direct Prisma access (requires DB) | Self-hosted, on-premise deployments |
Tools
Free Tools (5) — No API Key Required
| Tool | Description |
|------|-------------|
| detect_escalation | Analyze a message for Gottman-based escalation signals. Returns stage, score, signals, recommendations. EN+DE. |
| analyze_thread | Analyze conversation trajectory across 2+ messages. Returns score history, direction, velocity. |
| get_product_info | Product information — pricing, features, use cases, integration options. |
| get_api_documentation | API docs with code examples in curl, JS, Python. |
| get_psychology_foundations | Scientific foundations for conflict resolution and mediation. |
Premium Tools (12) — Requires API Key
Dialogue Management:
| Tool | Description |
|------|-------------|
| get_dialogue_types | List available dialogue types for conflict resolution. |
| analyze_conflict_type | Recommend the best dialogue type for a conflict situation. |
Mediation:
| Tool | Description |
|------|-------------|
| initiate_mediation | Start a mediation — creates dialogue, invites parties via email. |
| get_mediation_status | Check status and progress of an existing mediation. |
Agent Mediation Protocol (AI-to-AI negotiation):
| Tool | Description |
|------|-------------|
| start_agent_mediation | Start an XState-driven negotiation session (2-10 parties). |
| submit_position | Agent submits position for the current round. |
| get_proposals | Retrieve AI-generated proposals for consensus. |
| vote_on_proposal | Agent votes AGREE / DISAGREE / UNDECIDED on a proposal. |
| make_concession | Agent makes a tracked concession (yielding = strength). |
| get_session_status | Check current round, positions, and proposals. |
Scheduling:
| Tool | Description |
|------|-------------|
| schedule_dialogue_meeting | Propose meeting times, generates ICS calendar invite. |
| get_dialogue_schedule | Get scheduled meetings and attendee responses. |
Environment Variables
| Variable | Required | Description |
|----------|----------|-------------|
| PEACETHINKING_API_KEY | For premium tools | API key from peacethinking.ai (pt_live_*) |
| PEACETHINKING_API_URL | No | Custom API URL (default: https://peacethinking.ai) |
Part of the Peacethinking Ecosystem
@peacethinking/mcp-server is the AI agent interface of the Peacethinking Resolution Platform.
| Package | Description |
|---------|-------------|
| @peacethinking/conflict-framework | Shared types, schemas, and the 6-Dimensions Conflict Framework |
| @peacethinking/escalation-detector | Hybrid escalation detection — rule-based (OSS) + LLM-enhanced |
| @peacethinking/mediation-prompts | LLM prompt builders for verdict, perspective, and mediation flows |
| @peacethinking/mediation-workflows | XState-based mediation state machines and templates |
| @peacethinking/llm-eval | Benchmark any LLM on conflict resolution scenarios |
| @peacethinking/widget | Embeddable Web Component for escalation detection |
License
MIT — see LICENSE.
Built by peacethinking.ai.
