MCP servers de Docker

Connects AI clients to GitHub so they can read repositories, manage issues and pull requests, and inspect workflows.

Lets an AI client query and control an Unraid server's array, VMs, Docker containers and notifications via its GraphQL API.

Manages self-hosted cloud development environments and AI coding agents using Terraform-defined workspaces.

Lets an AI client manage Fly.io apps and machines, deploy Docker images, read logs, and run commands on machines.

Lets an AI client analyze repositories, generate Dockerfiles and Kubernetes manifests, build and scan images, and deploy containers.

Lets an AI client query and mutate Unraid NAS resources, including Docker containers, virtual machines, storage arrays, and notifications.

Provides AI assistants with tools for Arm architecture development, code migration from x86, and assembly performance analysis.

Lets an AI client manage Docker environments, Compose projects, containers, images, and GitOps syncs via Arcane.

Enables an AI client to create, manage, and operate Kubernetes clusters and workloads across local and cloud environments.

Lets AI assistants query and manage Kubernetes clusters by running kubectl commands from natural language requests.

Enables AI clients to manage Docker infrastructure, SSH hosts, ZFS, and file operations through specialized flux and scout tools.

Deploys and operates fleets of AI agents as production services with durable Postgres state, sandboxed execution, channels, and scheduled jobs.

Provides a self-hosted intelligence layer for homelab logs, metrics, and AI transcripts using SQLite and FTS5 search.

An open-source, self-hosted Slack alternative where human teams and AI agents collaborate in channels using MCP tools, memory, and shared browsers.

Lets an AI client delegate goals to a lead agent that plans and dispatches work to containerized worker agents with shared memory.

A self-hosted gateway that manages Codex-compatible agents and tools through stable API keys and automatic account routing.

Govern AI agent actions in Docker with deterministic policy, human approval, and cryptographically signed Decision Dossiers.

Lets developers connect AI applications to thousands of APIs, data sources, and tools through MCP with a single function call.

Securely execute LLM-generated code in isolated containers or local Linux and Windows sandboxes across multiple programming languages.

Provides AI assistants with deep semantic understanding of codebases through hybrid search, dependency graphs, and symbol-level impact analysis.

Exposes all Dokploy API endpoints as tools to allow AI clients to manage projects, applications, databases, and Docker configurations.

Lets an AI client query aggregated multi-platform trending news, RSS feeds, and AI-generated opinion analysis with sentiment and trend insights.

Lets AI clients run private web searches, read URLs, and get query suggestions through a SearXNG instance.

Lets AI clients decompile binaries, rename and type functions, run P-code emulation, and debug through Ghidra's reverse engineering engine.

Deploys applications and files to Google Cloud Run and lists services, projects, and service logs.

Enables AI agents to monitor Docker and Kubernetes infrastructure, investigate logs, and manage alerts through a built-in MCP server.

Lets an AI client log in to Xiaohongshu, publish image or video posts, search feeds, read post details and comments, and like, favorite, or reply to posts.

Provides a local-first runtime for AI agents with sandboxed tool execution, persistent sessions, and credential management.

Manage, deploy, and debug your self-hosted PaaS infrastructure using natural language through 46 specialized tools.

Lets AI agents run maigret OSINT searches to find where a username is registered across sites and analyze profile URLs.

Operate Docker-based YDB deployments through tools for lifecycle management, schema generation, health diagnostics, and authentication hardening.

Lets an AI client manage and search Sonarr, Radarr, Lidarr, and Prowlarr media libraries, downloads, and indexers.

Lets Claude clients chat with a self-hosted OpenClaw AI assistant, run async tasks, and route requests across multiple gateways.

Lets AI agents discover, load, and propose changes to a team's approved skills, tools, and company context stored in a git repository.

Drafts network allow lists for agent tools and runs them in a network-blocked sandbox to identify runtime connectivity issues.

Lets AI assistants inspect and manage Kubernetes clusters, debug pods, deploy workloads, and audit security via natural language.

Enables AI agents to browse the web, execute recorded playbooks, and allow human takeover through a HIPAA-compliant agentic browser.

Provides an AI agent and dashboard to monitor hosts, Docker containers, GPUs, and services across a self-hosted fleet via SSH.

Enables AI assistants to execute malware analysis tools and automated toolchains on a REMnux system via Docker, SSH, or HTTP.

Lets an AI assistant perform read-only full-text searches and triage across multiple IMAP email accounts via a local mirrored index.

Lets AI assistants build and edit goal, process and organisation maps on a self-hosted instance via a built-in MCP server.

Connects AI agents to Elasticsearch data so they can list indices, fetch mappings, and run queries in natural language.

Lets an AI client interact with an Unraid server's GraphQL API to manage arrays, VMs, Docker containers, and system notifications.

Wraps Docker CLI commands (ps, images, logs, build) so AI clients receive schema-validated typed JSON instead of raw terminal text.

Lets an AI client run Docker commands like ps, images, logs, build, and compose and receive typed, schema-validated JSON instead of raw terminal text.

Serve Obsidian Markdown vaults to browsers and AI agents, letting clients read, search, create, edit, move, and link notes.

Lets an AI client pull Scout APM endpoint metrics, traces, error groups, and N+1 insights to diagnose and fix performance problems.

Enables AI assistants to debug PHP applications using Xdebug's DBGp protocol through breakpoints, variable inspection, and execution control.

Lets an AI agent drive a self-hosted Chrome session that a human can watch and take over through a live noVNC view.

Provides remote access to a ChromaDB vector database via MCP, enabling shared semantic memory across Claude desktop, mobile, and CLI clients.