By jmagar
Drive the agent-os Windows 11 sandbox VM through its Windows-MCP server. Self-registers the windows-mcp MCP from your configured URL + bearer token, ships the agent-os skill, a /agent-os status command, and a SessionStart health check.
This plugin requires configuration values that are prompted when the plugin is enabled. Sensitive values are stored in your system keychain.
agent_os_mcp_urlFull HTTP(S) URL of the Windows-MCP endpoint inside the agent-os VM, including the /mcp path. This is what the plugin registers as the windows-mcp MCP server. Example: https://agent-os.example.ts.net/mcp. Prefer the VM's own stable Tailscale/MagicDNS name so it follows the VM across Docker hosts.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
${user_config.agent_os_mcp_url}agent_os_vm_hostSSH alias/host of the Docker host that runs the agent-os-win11 container (for `docker ps`, compose up, etc.). Example: tootie.
${user_config.agent_os_vm_host}agent_os_autostartSet to "true" to let the SessionStart hook SSH to the Docker host and `docker compose up -d` the VM when the Windows-MCP endpoint is unreachable. Requires agent_os_vm_host + agent_os_compose_file. It only starts (never stops) the VM and does not block on the multi-minute Windows boot. Leave "false" to only warn. Note: this starts an already-provisioned server; it does not install Windows-MCP from scratch.
${user_config.agent_os_autostart}agent_os_mcp_tokenBearer token the Windows-MCP server expects (sent as 'Authorization: Bearer <token>'). Stored in secure OS storage, not settings.json. Without it the windows-mcp server will return 401 and no tools will load.
${user_config.agent_os_mcp_token}agent_os_novnc_urlBase URL of the dockur/windows web (noVNC) UI on the Docker host, for visual debugging. Example: http://tootie:8006.
${user_config.agent_os_novnc_url}agent_os_compose_fileAbsolute path to the agent-os compose file ON the Docker host, used to bring the VM back up. Example: /mnt/cache/compose/windows/docker-compose.yml.
${user_config.agent_os_compose_file}agent_os_container_nameName of the dockur/windows container running the VM. Example: agent-os-win11.
${user_config.agent_os_container_name}agent_os_vm_tailscale_ipThe agent-os guest's own Tailscale IP (or MagicDNS name) used by `ssh agent-os` for in-guest commands on port 22. Example: 100.109.125.128.
${user_config.agent_os_vm_tailscale_ip}agent_os_host_forward_sshSSH user@host of the Docker host's port-forward into the guest (user@<docker-host-tailscale-ip>). Use this path — NOT `ssh agent-os` — for Tailscale maintenance inside the guest, since it survives `tailscale down`. Example: [email protected].
${user_config.agent_os_host_forward_ssh}agent_os_host_forward_portPort the Docker host forwards to the guest's sshd (host:PORT -> guest:22). Example: 2222.
${user_config.agent_os_host_forward_port}Dendrite is the Claude Code and Codex marketplace home for agent plugins, skills, MCP integrations, hooks, commands, and OpenAI agent companion files.
It is split out from jmagar/lab so the Lab control plane can stay focused on
the labby runtime while the agent capability catalog can evolve as its own
public repo.
.claude-plugin/marketplace.json: Claude Code marketplace manifest..agents/plugins/marketplace.json: Codex/OpenAI agent marketplace manifest.plugins/: local plugin sources carried by this repo.plugins/labby intentionally stays in jmagar/lab. The marketplace still
publishes labby as an external GitHub subdirectory source:
https://github.com/jmagar/lab.git, path plugins/labby.
| Plugin | Description | Skills | MCP servers | OpenAI agents | Commands |
|---|---|---|---|---|---|
acp | Agent Client Protocol skills for Rust, Python, and TypeScript implementations. | rust | none | 1 | none |
adguard | Skill for operating adguard via the lab MCP server / CLI. | adguard | none | 1 | none |
agent-os | Drive the agent-os Windows 11 sandbox VM through its Windows-MCP server. Self-registers the windows-mcp MCP from your configured URL + bearer token, ships the agent-os skill, a /agent-os status command, and a SessionStart health check. | agent-os | windows-mcp | 1 | agent-os.md |
arrs | The *arr / media-automation stack in one plugin: Radarr, Sonarr, Prowlarr, Overseerr, SABnzBD, qBittorrent, Plex, Jellyfin, Tautulli, and Tracearr, each operated via its own REST API. Credentials are configured here and bridged to the skills via a generated env file. | jellyfin, overseerr, tautulli, sabnzbd, radarr, tracearr, prowlarr, plex, qbittorrent, sonarr | none | 10 | none |
bitwarden | Bitwarden and secrets-management workflows for homelab operations. | bitwarden | bitwarden | 1 | bw-list.md, bw-get.md, bw-generate.md |
broadcastr | Helper assets for Broadcastr plugin tooling. | none | none | 0 | none |
bytestash | Skills for operating a ByteStash snippet manager. | bytestash | none | 1 | none |
dozzle | Skill for operating Dozzle through direct HTTP API checks, auth guidance, and MCP setup notes. | dozzle | dozzle | 1 | none |
immich | Skill for operating immich via the lab MCP server / CLI. | immich | none | 1 | none |
linkding | Skills for operating a Linkding bookmark manager. | linkding | none | 1 | none |
loggifly | Skill for operating loggifly via the lab MCP server / CLI. | loggifly | none | 1 | none |
memos | Skills for operating a Memos note hub. | memos | none | 1 | none |
navidrome | Operate a self-hosted Navidrome music server through its Subsonic API with direct HTTP calls: ping, browse artists/albums, search, and list playlists. | navidrome | none | 1 | none |
neo4j | Skill for operating neo4j via the lab MCP server / CLI. | neo4j | none | 1 | none |
notebooklm | NotebookLM research, source, generation, and download workflows. | notebooklm | none | 1 | none |
plexus | Remote-device memory and live operating context for host-specific work. | bootstrap-plexus, operating-remote | none | 2 | remote-context.md |
qdrant | Skill for operating qdrant via the lab MCP server / CLI. | qdrant | none | 1 | none |
radicale | CalDAV and CardDAV workflow skills for Radicale. | radicale | none | 1 | none |
scripts | Shared plugin maintenance scripts. | none | none | 0 | none |
scrutiny | Skill for operating scrutiny via the lab MCP server / CLI. | scrutiny | none | 1 | none |
swag | SWAG reverse proxy configuration management via MCP. Create, edit, view, and manage nginx proxy configurations with auth integration. | swag | swag-mcp, swag-mcp-remote | 1 | none |
tei | Skill for operating tei via the lab MCP server / CLI. | tei | none | 1 | none |
testing | App-testing and MCP-tooling skills: live QA of web, Android, and desktop apps; MCP server smoke-testing with mcporter; MCP-UI / Apps validation with mcpjam; and claude-in-mobile device automation. | mcpjam-ui-testing, android-app-testing, desktop-app-testing, web-app-testing, mcporter, claude-in-mobile | none | 6 | none |
uptime-kuma | Read-only monitoring of a self-hosted Uptime Kuma instance via direct HTTP: Prometheus /metrics with API-key auth and public status-page JSON. No monitor management. | uptime-kuma | none | 1 | none |
npx claudepluginhub jmagar/dendrite --plugin agent-osQuery, monitor, and manage Unraid servers via GraphQL API through MCP tools. Supports system info, Docker, VMs, array/parity, notifications, plugins, rclone, and live telemetry.
Core homelab agents, commands, and setup/health skills for self-hosted service management. Includes interactive credential setup wizard and unified service health dashboard.
UniFi network management via MCP tools. Monitor devices, clients, network health, firewall rules, and perform management operations.
Gotify push notifications and management via MCP tools with HTTP fallback. Sends alerts for long-running tasks, plan completions, and blocked states.
Agents, commands, skills, and scripts for scaffolding, reviewing, aligning, and deploying homelab MCP server plugins. Includes canonical Python, TypeScript, and Rust server templates.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development