Arc Gate — LLM 代理工具 是 AI Skill Hub 本期精选MCP工具之一。综合评分 7.5 分,整体质量较高。我们推荐使用将其纳入你的 AI 工具库,帮助提升工作效率。
Arc Gate — LLM 代理工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
Arc Gate — LLM 代理工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/9hannahnine-jpg/arc-gate
# 方式二:手动配置 claude_desktop_config.json
{
"mcpServers": {
"arc-gate---llm-----": {
"command": "npx",
"args": ["-y", "arc-gate"]
}
}
}
# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
# 安装后在 Claude 对话中直接使用 # 示例: 用户: 请帮我用 Arc Gate — LLM 代理工具 执行以下任务... Claude: [自动调用 Arc Gate — LLM 代理工具 MCP 工具处理请求] # 查看可用工具列表 # 在 Claude 中输入:"列出所有可用的 MCP 工具"
// claude_desktop_config.json 配置示例
{
"mcpServers": {
"arc_gate___llm_____": {
"command": "npx",
"args": ["-y", "arc-gate"],
"env": {
// "API_KEY": "your-api-key-here"
}
}
}
}
// 保存后重启 Claude Desktop 生效
The runtime governance proxy for AI agents — part of the Bendex Arc platform.
Catches multi-turn attacks that unfold across a conversation, not just single messages.
1. Fork this repo 2. Create a Railway project from the fork 3. Set environment variables: - OPENAI_API_KEY — your OpenAI key - GATE_BLOCK_MODE — true - GATE_UPSTREAM — https://api.openai.com - GATE_BASE_URL — your Railway URL 4. Railway auto-deploys from the Procfile
Arc Gate ships with prebuilt runtime governance policies for common agent deployments. Set the policy at deployment time via environment variable or per-request via header.
Environment variable (applies to all requests):
ARC_POLICY_MODE=finance_agent
Per-request header (overrides environment):
x-arc-policy-mode: finance_agent
Available templates:
browser_agent — For browser and web automation agents. Webpages and external content treated as untrusted. External actions blocked under ambiguity. Read-only continuation allowed.
finance_agent — For financial agents handling payments, transfers, and account data. Strictest defaults. Payment and transfer actions restricted under any elevated risk. Analysis allowed, transactions require clean session.
rag_assistant — For RAG pipelines and document retrieval systems. Retrieved documents are informational only and cannot issue instructions. Safe summarization preserved. No tool or workflow escalation from retrieved content.
balanced — Default. Recommended for most deployments.
strict — Maximum protection. Higher false positive rate. For high-risk deployments.
research — Reduced blocking for security research and red-teaming.
developer — Minimal blocking for development and testing. Not for production.
See the attack in action — same LangChain agent, with and without Arc Gate:
See the attack in action — same LangChain agent, with and without Arc Gate:
| Framework | Package | Install |
|---|---|---|
| LangChain | [langchain-arcgate](https://pypi.org/project/langchain-arcgate/) | pip install langchain-arcgate |
| CrewAI | [crewai-arcgate](https://pypi.org/project/crewai-arcgate/) | pip install crewai-arcgate |
| MCP | [arc-gate-mcp](https://pypi.org/project/arc-gate-mcp/) | pip install arc-gate-mcp |
该项目提供了开源的MCP工具,支持LLM代理与提示注入检测,适合AI安全和代理开发,但代码质量和文档完善度需要进一步提高
该工具使用 NOASSERTION 协议,商用场景请仔细阅读协议条款,必要时咨询法律意见。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
📄 NOASSERTION — 请查阅原始协议条款了解具体使用限制。
经综合评估,Arc Gate — LLM 代理工具 在MCP工具赛道中表现稳健,质量良好。如果你已有明确的使用需求,可以直接上手体验;如果还在评估阶段,建议对比同类工具后再做决策。
| 原始名称 | arc-gate |
| 原始描述 | 开源MCP工具:Arc Gate — LLM proxy with prompt injection detection. Bendex Geometry.。⭐7 · HTML |
| Topics | mcpagent-securityai-agentsai-securitycrewailangchain |
| GitHub | https://github.com/9hannahnine-jpg/arc-gate |
| License | NOASSERTION |
| 语言 | HTML |
收录时间:2026-06-13 · 更新时间:2026-06-13 · License:NOASSERTION · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端