A
licenseA
qualityC
maintenanceLet Claude and Cursor manage your Railway infrastructure through natural language. Deploy, configure, and monitor - autonomously and safely.
38
97 npm
72
MIT
Lizard MCP
Officialby lizard-build
Related Servers
Alternatives to Lizard MCP
PlugLayer MCP Serverofficial
AlicenseCqualityBmaintenanceEnables deploying and managing infrastructure via natural language, including project/domain management, compute nodes, image deployment, and CI/CD integration.70MIT- AlicenseNot gradedqualityCmaintenanceEnables managing Sliplane deployments, projects, and application status through natural language, with OAuth or API key auth.MIT
Related Servers
hostman-mcp-serverofficial
AlicenseAqualityDmaintenanceEnables deploying applications to Hostman, creating VPCs, databases, and floating IPs, and managing VCS providers through natural language prompts.126 npmMIT- AlicenseAqualityDmaintenanceEnables managing DigitalOcean cloud resources, including Droplets, databases, domains, and more, via natural language.121MIT
- AlicenseBqualityDmaintenanceEnables management of Komodo infrastructure including servers, stacks, deployments, builds, repositories, procedures, and system settings through natural language.5496 npm10GPL 3.0
- AlicenseNot gradedqualityNot gradedmaintenanceEnables AI assistants to manage cloud infrastructure through natural language by providing a unified interface to the Dokploy platform. Supports Docker containers, applications, databases, domains, monitoring, and deployment operations through conversational commands.10 npm1-
- AlicenseAqualityBmaintenanceEnables AI assistants to deploy and manage SnapDeploy containers and resources—including builds, logs, environment variables, databases, and custom domains—directly through natural language.1511 npmMIT
- AlicenseAqualityDmaintenanceEnables deployment and management of applications on the Citizen platform from git repositories or local files. It allows AI assistants to monitor deployment logs, handle authentication, and automatically fix build errors using intelligent error analysis.112 npmMIT
TDQS
A3.6/5.0
Scored across 56 tools
Disambiguation4/5
Tool purposes are mostly distinct, with clear resource/action prefixes such as service_pause vs project_pause vs workspace_pause, and deploy_redeploy vs deploy_restart. A few boundaries are softer — service_set vs service_scale vs service_set_port/config_apply, billing_status vs billing_summary, and sandbox_exec vs ssh_exec — but the detailed descriptions help differentiate them.
Naming Consistency4/5
Names consistently use snake_case and a resource_action pattern, making the set predictable. Minor deviations like whoami, billing_summary/billing_status, and service_sleep_status/service_sleep_set break the verb-action pattern slightly but remain readable and coherent.
Tool Count2/5
56 tools is excessive for a single MCP server; many granular operations (service_get_port, service_set_port, service_sleep_status, service_sleep_set) could be folded into broader tools. While the platform is broad, the count far exceeds typical coherence thresholds (3–15) and makes discovery and selection costly.
Completeness4/5
Core lifecycle coverage is strong for services, secrets, domains, deploys, sandboxes, and templates, with config_apply filling bulk-operation gaps. Minor gaps remain — e.g., no workspace_delete, no explicit volume_create/delete/attach, and limited addon-specific update tools — but agents can likely work around them.
Maintenance
ActivityActive
ResponsivenessUnresponsive