AgentBreakWatch

Audit AI Agent SDKs and MCP servers for breaking changes, schema drift, and migration fixes.

Should I use this

Quality & Safety

A
Description quality
100%
Schema completeness
80%
Naming quality
100%
Poisoning risk
100%
Permission match
100%
Protocol compliance
100%

Based on automated analysis of tool definitions and protocol compliance.

Context Cost

~692Tokens (tool definitions)
~736 BTypical response size
Moderate attention impact (0.54% of 128k context)

This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.

Install

One-Click Install

Add this to your `claude_desktop_config.json` file:

{
  "mcpServers": {
    "agentbreakwatch-mcp": {
      "url": "https://agentbreakwatch.com/api/mcp"
    }
  }
}

Remote endpoints

https://agentbreakwatch.com/api/mcpstreamable-http

What it can do

Tool inventory

Tools (4)

🟢 Read-only🟡 Write🔴 Delete⚪ Unknown
🟢check_stack_drift(manifest)

Read-only audit of a package.json, requirements.txt, or .mcp.json configuration string against 256 live AI Agent SDKs and MCP servers. Flags breaking major version deltas, unpinned @latest MCP servers, and outdated dependencies with exact pin commands. Requires no authentication and returns a structured JSON report.

Input Schema

{
  "type": "object",
  "properties": {
    "manifest": {
      "type": "string",
      "description": "Raw contents of .mcp.json, package.json, requirements.txt, or a newline-separated list of packages"
    }
  },
  "required": [
    "manifest"
  ],
  "additionalProperties": false
}
🟢get_package_stability(package)

Read-only lookup of live npm or PyPI version, publish timestamp, 90-day release velocity, stability score (0-100), MCP tools/list schema drift risk, and before/after migration code for a single AI Agent SDK or MCP server. Prefer this over check_stack_drift when inspecting one specific package by name or slug.

Input Schema

{
  "type": "object",
  "properties": {
    "package": {
      "type": "string",
      "description": "Package name or slug (e.g. 'openai', '@modelcontextprotocol/sdk', '@playwright/mcp', 'fastmcp', 'mcp-clickhouse')"
    }
  },
  "required": [
    "package"
  ],
  "additionalProperties": false
}
🟢list_volatile_packages(kind, category, limit)

Read-only query that lists the fastest-changing or highest schema-drift-risk AI Agent SDKs and MCP servers from the live AgentBreakWatch telemetry index, sorted descending by 90-day release velocity. Supports filtering by kind ('sdk', 'mcp', 'all'), category hub slug, and result limit (1-50).

Input Schema

{
  "type": "object",
  "properties": {
    "kind": {
      "type": "string",
      "enum": [
        "all",
        "sdk",
        "mcp"
      ],
      "description": "Filter by Agent SDK ('sdk'), MCP server ('mcp'), or 'all'"
    },
    "category": {
      "type": "string",
      "description": "Optional category hub slug (e.g. 'agent-sdks', 'aws-labs', 'database-backend', 'browser-automation', 'cloud-infra')"
    },
    "limit": {
      "type": "number",
      "description": "Maximum number of packages to return (default 15, max 50)"
    }
  },
  "additionalProperties": false
}
🟢get_archived_mcp_replacements

Read-only lookup returning the complete migration matrix for archived or frozen @modelcontextprotocol/server-* reference packages and their official 2026 vendor-maintained replacements. Takes no parameters and requires no authentication.

Input Schema

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}

Community

Rate this Server

Evidence

Recent observations

verifiedversion not recorded4 tools