x=hub

your toolset

    build deployment
    sign in

    microsoft-planner-mcp

    projectother scanned
    io.github.aixolotl/microsoft-planner-mcp · v
    deploy just this

    Unofficial MCP server for Microsoft Planner using Microsoft Graph and Entra ID delegated authentication

    run it

    on your own machines, one way per deploy method
    generated from the record, not tested yet
    CHANGE_ME: built from the repository's Dockerfile, assuming it serves streamable HTTP at :8000/mcp. Nothing has run this yet; the worker replaces it with a tested recipe.
    # once, as root:
    docker build -t mcp-local/microsoft-planner:main 'https://github.com/aixolotl/microsoft-planner-mcp.git#main'
    
    #!/usr/bin/env bash
    # microsoft-planner (io.github.aixolotl/microsoft-planner-mcp), recipe docker (generated).
    # Generated by Hub for toolset "my-agent"; started by mcp-my-agent-microsoft-planner.service.
    set -euo pipefail
    export HOST=127.0.0.1 PORT=8101
    exec docker run --rm --name mcp-my-agent-microsoft-planner -p 127.0.0.1:${PORT}:8000 mcp-local/microsoft-planner:main
    
    [Unit]
    Description=MCP server microsoft-planner (toolset my-agent), generated by Hub
    After=network-online.target docker.service
    Wants=network-online.target
    
    [Service]
    Type=simple
    User=mcp
    Group=mcp
    SupplementaryGroups=docker
    EnvironmentFile=/etc/mcp/my-agent.env
    Environment=HOME=/var/lib/mcp/microsoft-planner
    WorkingDirectory=/var/lib/mcp/microsoft-planner
    ExecStart=/opt/mcp/my-agent/bin/microsoft-planner.sh
    Restart=on-failure
    RestartSec=5
    NoNewPrivileges=true
    ProtectSystem=strict
    ProtectHome=true
    PrivateTmp=true
    ReadWritePaths=/var/lib/mcp/microsoft-planner
    
    [Install]
    WantedBy=multi-user.target
    
    generated from the record, not tested yet
    CHANGE_ME: built from the repository's Dockerfile, assuming it serves streamable HTTP at :8000/mcp. Nothing has run this yet; the worker replaces it with a tested recipe.
    docker build -t microsoft-planner 'https://github.com/aixolotl/microsoft-planner-mcp.git#main'
    docker run --rm -p 127.0.0.1:8101:8000 microsoft-planner
    
    generated from the record, not tested yet
    CHANGE_ME: built from the repository's Dockerfile, assuming it serves streamable HTTP at :8000/mcp. Nothing has run this yet; the worker replaces it with a tested recipe.
    # Generated by Hub for toolset "my-agent" (2026-10-19T15:02:43+00:00).
    # Each package server runs behind a stdio->HTTP bridge and listens on localhost only.
    #   cp .env.example .env    # fill the ${VAR}s
    #   docker compose up -d && bash smoke.sh
    # Remote servers are not here: clients reach them directly (see clients/).
    name: mcp-my-agent
    services:
      microsoft-planner:
        build:
          context: https://github.com/aixolotl/microsoft-planner-mcp.git#main
        ports:
        - 127.0.0.1:8101:8000
        restart: unless-stopped
        x-note: 'CHANGE_ME: built from the repository''s Dockerfile, assuming it serves streamable HTTP at
          :8000/mcp. Nothing has run this yet; the worker replaces it with a tested recipe.'
    volumes:
      npm-cache: {}
      uv-cache: {}
    
    generated from the record, not tested yet
    CHANGE_ME: built from the repository's Dockerfile, assuming it serves streamable HTTP at :8000/mcp. Nothing has run this yet; the worker replaces it with a tested recipe.
    # Generated by Hub for toolset "my-agent" (2026-10-19T15:02:43+00:00).
    #   kubectl create namespace mcp
    #   kubectl -n mcp create secret generic mcp-credentials --from-env-file=.env
    #   kubectl apply -f kubernetes.yaml
    # Services are cluster-internal: <name>.mcp.svc.cluster.local:8000/mcp. Put an ingress
    # with auth in front before exposing any of them.
    apiVersion: v1
    kind: Namespace
    metadata:
      name: mcp
    ---
    apiVersion: apps/v1
    kind: Deployment
    metadata:
      name: microsoft-planner
      namespace: mcp
      labels: &id001
        app: microsoft-planner
    spec:
      replicas: 1
      selector:
        matchLabels: *id001
      template:
        metadata:
          labels: *id001
        spec:
          containers:
          - name: microsoft-planner
            image: CHANGE_ME
            ports:
            - containerPort: 8000
              name: http
            env: []
    ---
    apiVersion: v1
    kind: Service
    metadata:
      name: microsoft-planner
      namespace: mcp
    spec:
      selector:
        app: microsoft-planner
      ports:
      - port: 8000
        targetPort: http
        name: http
    
    generated from the record, not tested yet
    CHANGE_ME: built from the repository's Dockerfile, assuming it serves streamable HTTP at :8000/mcp. Nothing has run this yet; the worker replaces it with a tested recipe.
    # Generated by Hub for toolset "my-agent" (2026-10-19T15:02:43+00:00).
    # Credentials come from the Secret 'mcp-credentials' (kubectl create secret generic ... --from-env-file=.env).
    # Each server under `servers` is one Deployment + Service; the keys after it configure
    # upstream charts a worker verified.
    servers:
      microsoft-planner:
        enabled: true
        replicas: 1
        port: 8000
        path: /mcp
        container:
          name: microsoft-planner
          image: CHANGE_ME
          ports:
          - containerPort: 8000
            name: http
          env: []
    
    # helm install my-agent helm/my-agent -n mcp
    

    Scanned github.com/aixolotl/microsoft-planner-mcp at 4c13b98f8389 with [email protected][email protected] on 2026-10-19 before the build: 14 dependencies, 62 known vulnerabilities (2 critical, 30 high, 26 medium, 4 low), nothing in the code that looks malicious.

    in the cloud

    one click: the provider's deploy form opens filled in, in your own account
    koyeb

    Clients connect to the service's https://…koyeb.app address plus /mcp.

    cloud run

    Cloud Shell clones the repository, builds its Dockerfile and deploys it to Cloud Run.

    Cloud Run sends requests to $PORT (8080); the recipe expects :8000, so the server has to read PORT, or set the container port on the service afterwards.

    connect to it

    copy what your client needs
    {
      "mcpServers": {
        "microsoft-planner": {
          "command": "CHANGE_ME",
          "args": []
        }
      }
    }
    #!/usr/bin/env bash
    # Registers every server with Claude Code (run once; `claude mcp list` to check).
    set -euo pipefail
    set -a; [ -f .env ] && source .env; set +a
    claude mcp add microsoft-planner -- CHANGE_ME
    
    {
      "mcpServers": {
        "microsoft-planner": {
          "command": "CHANGE_ME",
          "args": []
        }
      }
    }
    {
      "servers": {
        "microsoft-planner": {
          "type": "stdio",
          "command": "CHANGE_ME",
          "args": []
        }
      }
    }
    name: my-agent
    target: local
    servers:
    - hub: io.github.aixolotl/microsoft-planner-mcp
    
    • not probed yet: tools unpinned, ingest lists them
    • no tested recipe for target 'local' yet: build it from https://github.com/aixolotl/microsoft-planner-mcp by hand, or wait for the worker

    tools

    Probe result: faileddocker: the container exited: …[level:], package, level) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "<frozen importlib._bootstrap>", line 1387, in _gcd_import File "<frozen importlib._bootstrap>", line 1360, in _find_and_load File "<frozen importlib._bootstrap>", line 1331, in _find_and_load_unlocked File "<frozen importlib._bootstrap>", line 935, in _load_unlocked File "<frozen importlib._bootstrap_external>", line 999, in exec_module File "<frozen importlib._bootstrap>", line 488, in _call_with_frames_removed File "/app/src/server.py", line 24, in <module> from .auth_provider import auth File "/app/src/auth_provider.py", line 18, in <module> from .config import settings File "/app/src/config.py", line 131, in <module> settings = Settings() # ty:ignore[call-arg] ^^^^^^^^^^ File "/app/.venv/lib/python3.12/site-packages/pydantic_settings/main.py", line 242, in __init__ super().__init__(**__pydantic_self__.__class__._settings_build_values(sources, init_kwargs)) File "/app/.venv/lib/python3.12/site-packages/pydantic/main.py", line 250, in __init__ validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ pydantic_core._pydantic_core.ValidationError: 3 validation errors for Settings CLIENT_ID Field required [type=missing] CLIENT_SECRET Field required [type=missing] TENANT_ID Field required [type=missing]

    runs · 1

    the latest tests of this entry by the hub and its workers, newest first
    dateworkermethodstatussandboxreason
    2026-10-19 09:50 local-mac-vm docker failed gvisor docker: the container exited: …[level:], package, level) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "<frozen importlib._bootstrap>", line 1387, i…