Skip to main content
Glama
README.md
# ORACLE-H MCP Server

> Human-on-the-Loop validation for autonomous AI agents — via Model Context Protocol

[![npm version](https://badge.fury.io/js/oracle-h-mcp.svg)](https://www.npmjs.com/package/oracle-h-mcp)
[![npm downloads](https://img.shields.io/npm/dm/oracle-h-mcp.svg)](https://www.npmjs.com/package/oracle-h-mcp)
[![smithery badge](https://smithery.ai/badge/@tponscr-debug/oracle-h-mcp)](https://smithery.ai/server/@tponscr-debug/oracle-h-mcp)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

## What is ORACLE-H?

ORACLE-H is a safety infrastructure that acts as a mandatory checkpoint for autonomous AI agents. Before executing any **critical, destructive, or irreversible action**, the agent submits it for human approval via **Telegram**. The human decides with a single tap.

```
Agent → oracle_validate → Human gets Telegram alert → Approve/Reject → Agent proceeds
```

**Without ORACLE-H**: 4-step workflow succeeds ~81.5% of the time.
**With ORACLE-H**: Same workflow succeeds ~99.6% of the time.

## MCP Tools

| Tool | Description |
|------|-------------|
| `oracle_validate` | Submit a risky action for human approval. Returns a `request_id`. |
| `oracle_poll_status` | Poll the decision until approved, rejected, or expired. |

## Quick Start (Demo — no signup required)

```bash
# Test immediately with the public demo key
ORACLE_SIGNATURE_KEY=demo-signature-key-change-in-production
ORACLE_BASE_URL=https://oracle-h.online
```

---

## Installation

### Claude Desktop

Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "oracle-h": {
      "command": "npx",
      "args": ["-y", "oracle-h-mcp"],
      "env": {
        "ORACLE_SIGNATURE_KEY": "<your-signature-key>",
        "ORACLE_BASE_URL": "https://oracle-h.online"
      }
    }
  }
}
```

### Cursor / Windsurf

Add to your MCP settings:
```json
{
  "oracle-h": {
    "command": "npx",
    "args": ["-y", "oracle-h-mcp"],
    "env": {
      "ORACLE_SIGNATURE_KEY": "<your-signature-key>",
      "ORACLE_BASE_URL": "https://oracle-h.online"
    }
  }
}
```

### Cline (VS Code)

```json
// File: cline_mcp_settings.json
{
  "mcpServers": {
    "oracle-h": {
      "command": "npx",
      "args": ["-y", "oracle-h-mcp"],
      "env": {
        "ORACLE_SIGNATURE_KEY": "demo-signature-key-change-in-production",
        "ORACLE_BASE_URL": "https://oracle-h.online"
      }
    }
  }
}
```

### Roo Code (VS Code)

```json
// File: .roo/mcp.json (project-level)
{
  "mcpServers": {
    "oracle-h": {
      "command": "npx",
      "args": ["-y", "oracle-h-mcp"],
      "env": {
        "ORACLE_SIGNATURE_KEY": "demo-signature-key-change-in-production",
        "ORACLE_BASE_URL": "https://oracle-h.online"
      }
    }
  }
}
```

### GitHub Copilot (VS Code)

Add to VS Code settings.json:
```json
{
  "github.copilot.chat.mcp.servers": {
    "oracle-h": {
      "command": "npx",
      "args": ["-y", "oracle-h-mcp"],
      "env": {
        "ORACLE_SIGNATURE_KEY": "demo-signature-key-change-in-production",
        "ORACLE_BASE_URL": "https://oracle-h.online"
      }
    }
  }
}
```

### n8n

Use the MCP Client Tool node in n8n, running via supergateway:
```bash
npx -y supergateway --stdio "npx -y oracle-h-mcp" --port 8811
```
Then connect n8n's MCP Client Tool to http://localhost:8811/sse

### Zed Editor

Add to `settings.json`:
```json
{
  "context_servers": {
    "oracle-h": {
      "command": {
        "path": "npx",
        "args": ["-y", "oracle-h-mcp"]
      },
      "env": {
        "ORACLE_SIGNATURE_KEY": "<your-signature-key>",
        "ORACLE_BASE_URL": "https://oracle-h.online"
      }
    }
  }
}
```

### Continue.dev

Create `.continue/mcpServers/oracle-h.yaml`:
```yaml
name: oracle-h
version: 1.0.0
schema: v1
mcpServers:
  - name: oracle-h
    command: npx
    args:
      - -y
      - oracle-h-mcp
    env:
      ORACLE_SIGNATURE_KEY: <your-signature-key>
      ORACLE_BASE_URL: https://oracle-h.online
```

### LangGraph (Python)

```python
from langchain_mcp_adapters.client import MultiServerMCPClient

async with MultiServerMCPClient({
    "oracle-h": {
        "command": "npx",
        "args": ["-y", "oracle-h-mcp"],
        "env": {
            "ORACLE_SIGNATURE_KEY": "your-key",
            "ORACLE_BASE_URL": "https://oracle-h.online"
        },
        "transport": "stdio"
    }
}) as client:
    tools = client.get_tools()
```

---

## How it works

1. **Agent calls `oracle_validate`** with action, rationale, projected impact, and TTL
2. **Human receives a Telegram notification** with full context and risk analysis
3. **Human taps Approve or Reject** directly in Telegram
4. **Agent polls `oracle_poll_status`** until decision arrives
5. **Agent proceeds or aborts** based on the human decision

The system includes:
- 🛡️ **Anti-hallucination Watchdog** — detects context truncation before submitting
- 📊 **Reliability Calculator** — shows the statistical improvement (81.5% → 99.6%)
- ⏳ **TTL / Expiry** — requests auto-expire if no human response within the configured timeout
- 🔐 **Bearer Token Auth** — each agent has a unique signature key
- 🔒 **HMAC-SHA256 signed webhooks** — tamper-proof decisions

---

## System Prompt Snippet

Add this to your agent's system prompt to ensure ORACLE-H is always used correctly:

```
## Human Approval Gate — ORACLE-H (MANDATORY)

Call oracle_validate BEFORE executing any action in these categories:

DESTRUCTIVE: Deleting files, dropping databases, removing users, git force-push
EXTERNAL WRITES: Sending emails, posting to APIs, submitting forms
FINANCIAL: Payments, billing changes, subscription modifications
INFRASTRUCTURE: Deployments, server config, DNS, environment variables

POLLING PROTOCOL:
After oracle_validate, poll oracle_poll_status every 10-15 seconds until:
- "approved" → proceed
- "rejected" → STOP, inform user, propose alternatives
- "expired"  → STOP, ask user if they want to retry

HARD RULES:
- NEVER execute before "approved"
- If oracle_validate is unreachable: BLOCK the action, inform user
- Read-only operations do NOT require oracle_validate
```

---

## API

Full API documentation: [oracle-h.online/llms.txt](https://oracle-h.online/llms.txt)
OpenAPI spec: [oracle-h.online/openapi.json](https://oracle-h.online/openapi.json)

## License

MIT

TDQS

A4.2/5.0

Scored across 2 tools

Disambiguation5/5

The two tools have completely distinct and non-overlapping purposes: oracle_validate is for submitting actions for human approval, while oracle_poll_status is for checking the status of those submissions. Their descriptions clearly differentiate them, with no ambiguity in functionality.

Naming Consistency5/5

Both tools follow a consistent 'oracle_verb_noun' pattern (oracle_poll_status and oracle_validate), using snake_case and starting with the server prefix. This makes them predictable and easy to identify within the server's namespace.

Tool Count2/5

With only two tools, the server feels thin for its apparent purpose of handling human approval workflows. While the tools cover the core submit-and-poll cycle, there are likely missing operations like listing pending requests or managing notifications, making the scope underdeveloped.

Completeness3/5

The server provides basic coverage for the human approval domain with submit and poll operations, but there are notable gaps. Missing tools for operations like canceling requests, viewing request history, or configuring approval settings limit its completeness, though agents can still work with the core workflow.