mcp-sentinel

GCS-ZHN/mcp-sentinel

★ 3 stars TypeScript AI/LLM Updated today

Harness agent plugin that acts as a sentinel between the AI agent and MCP servers — polling long-running tasks so token-costly status loops never enter the LLM inference path

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "mcp-sentinel": {
      "command": "npx",
      "args": [
        "-y",
        "GCS-ZHN/mcp-sentinel"
      ]
    }
  }
}

README Excerpt

A **sentinel** between an AI agent and MCP servers — polling long-running tasks on the agent's behalf so that token-costly status loops never enter the LLM inference path. This is a **monorepo**: a harness-agnostic core (`@gcszhn/mcp-sentinel-core`) plus one thin plugin package per agent host. **Zero MCP re-configuration.** The sentinel never asks the user to configure MCP

Tools (14)

SENTINEL_MAX_POLL_LOGSENTINEL_TASK_TTL_MSactionargscontainsgteintervallimitltematchmcp_sentinel_polloffsettimeoutuntil

Topics

dsh-pluginopencode-plugin

mcp-sentinel

Quick Install

README Excerpt

Tools (14)

Topics

Related AI/LLM Servers