Developer releases 8 open-source MCP servers for AI-driven infra with built-in safety controls
A developer has published a suite of eight open-source Model Context Protocol (MCP) servers designed to give AI agents controlled access to production infrastructure. The servers cover tools spanning Kubernetes, Kafka, ClickHouse, Azure, Oracle Cloud, Debezium, Azure DevOps, and Keycloak. All eight share a common governance model featuring tiered access modes, allowlists, protected resource locks, destructive-action flags, and typed confirmation prompts that require echoing the exact target name before high-impact operations proceed. Credentials are automatically redacted before reaching the model, and every guarded operation emits a JSON audit log entry. Published under the MIT license as TypeScript packages on npm under the @dockndevai namespace, the servers are compatible with Claude, Cursor, Codex, VS Code, and Windsurf.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in