広告掲載
MCP.so
ログイン
P

Preloop MCP Server

@preloop

リポジトリ

基本情報

カテゴリ

その他

トランスポート

stdio

認証方式

oauth

作者

preloop

投稿者

Dimitris Moraitis

Preloop MCP Server について

Preloop is the MCP Governance Layer. Add approval gates to AI tool calls. Intercept risky actions, collect human decisions, and keep an audit trail.

#governance#automation#claude#cursor#audit#security#human-approval#gate

Preloop MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "preloop": {
      "url": "https://preloop.ai/mcp/v1",
      "transport": "http-streaming",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY_HERE"
      }
    }
  }
}

Preloop MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

このサーバーは認証が必要です。ツールの取得にはログインや提供元の認証情報が必要な場合があります。

上のボタンでサーバーのエンドポイントに接続し、ツール一覧を取得できます。

概要

What is Preloop?

Preloop is an open-source, event-driven automation platform with built-in human-in-the-loop safety. It acts as an MCP proxy, intercepting sensitive operations called by AI agents and routing them for human approval. It integrates with issue trackers like Jira, GitHub, and GitLab.

How to use Preloop?

Clone the repository, create a Python virtual environment (3.11+), install dependencies, set up PostgreSQL 14+ with PGVector, and configure environment variables in .env. Start the API server with ./start.sh and the sync service with preloop-sync scan all. Alternatively, deploy with Docker Compose or a Helm chart for Kubernetes.

Key features of Preloop

  • Event-driven automation with AI agents responding to tool events
  • Human-in-the-loop safety intercepting sensitive operations for approval
  • Built-in MCP server with 6 tools and JWT authentication
  • Tool management and proxying of external MCP servers
  • Issue tracker integration (Jira, GitHub, GitLab) with continuous sync
  • Vector search and automated duplicate detection
  • Compliance metrics and improvement recommendations
  • Modern web UI built with Lit, Vite, and Shoelace Web Components

Use cases of Preloop

  • Automate issue tracking workflows with human approval gates
  • Implement compliance checks and recommendations for issue descriptions
  • Detect duplicate and overlapping issues automatically
  • Enable AI agents to respond to events with safety oversight
  • Approve sensitive tool calls via email or mobile app notifications

よくある質問

1

What runtime dependencies are required?

Python 3.11+, PostgreSQL 14+ with the PGVector extension for vector search capabilities.

2

Can I disable self-registration?

Yes. Set `REGISTRATION_ENABLED=false` in your `.env` file or Docker environment. This hides the sign-up UI, redirects `/register` to `/login`, and returns 403 on the registration API endpoint.

3

How do I access the API documentation?

When the server is running, the interactive API docs are at `http://localhost:8000/docs` and the OpenAPI spec at `http://localhost:8000/openapi.json`.

4

What issue trackers are supported?

Jira (Cloud and Server), GitHub Issues, and GitLab Issues. Azure DevOps and Linear are planned for future releases.

5

What is the difference between Open Source and Enterprise Preloop?

The open-source version provides single-user approval workflows with email and mobile app notifications. The Enterprise edition adds RBAC, team-based approvals with quorum, escalation, and Slack/Mattermost notifications.

コメント

「その他」の他のコンテンツ

Awesome Mlops

visenger

A curated list of references for MLOps

14.0K

Blender

ahujasid

Open-source MCP to use Blender with any LLM

23.6K

Sequential Thinking

modelcontextprotocol

Model Context Protocol Servers

88.2K

Volley.sh

Volley

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi

MaxKB

1Panel-dev

🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。

22.0K
広告掲載