MCP Observatory: Scan, Test, and Secure MCP Servers Before Agents Depend on Them

MCP servers are becoming production dependencies.

If an agent depends on a server, that server needs the same basic safety net as any other production integration: CI checks, security review, regression reports, and artifacts that teams can inspect when behavior changes.

MCP Observatory helps teams test, secure, and monitor MCP servers before agents depend on them.

It can:

• scan MCP servers from local configs
• run in CI through the GitHub Action
• detect capability, schema, and response drift
• record, replay, and verify MCP sessions
• produce health scores, badges, and production/security reports
• run as an MCP server so agents can test other MCP servers

The open source CLI and MCP server remain free for local OSS use under MIT.

For teams running MCP in private repos or production workflows, paid pilots are available for hosted reporting, private repo CI history, security reports, production monitoring, certification, support, and fleet visibility.

Try it

npx @kryptosai/mcp-observatory@latest

Enter fullscreen mode Exit fullscreen mode

Deep scan:

npx @kryptosai/mcp-observatory@latest scan deep --security

Enter fullscreen mode Exit fullscreen mode

Running MCP in production? Contact [email protected].