Skip to main content
Glama
happypod-mcp

happypod-mcp

Official
by happy-pod

Happy Pod MCP server and skills

Use Happy Pod from your AI coding tool: ship an app, add a database, start a GPU pod, find out why production is slow and fix it, all by asking in plain language.

  • MCP server (happypod-mcp): 33 tools over the Happy Pod API. Works in Claude Code, Codex, Cursor, VS Code, Windsurf, Claude Desktop and any MCP client.

  • Skills (skills/): playbooks so the agent works the way Happy Pod expects (price first, confirm before deleting, read the logs when a deploy fails):

    • happypod: access, safety rules, which skill to use.

    • happypod-ship: app pods from a repo or image, managed Postgres and Redis, custom domains, GPU pods.

    • happypod-ops: diagnose, logs and metrics, roll back, restart, scale, keep the bill low.

Install

Claude Code (plugin: server and skills together)

/plugin marketplace add happy-pod/mcp
/plugin install happypod@happypod

Then give it your key (Settings → API keys at happypod.io) before starting Claude Code:

export HAPPYPOD_API_KEY=hp_...

Any MCP client (local server)

git clone https://github.com/happy-pod/mcp happypod-mcp
{
  "mcpServers": {
    "happypod": {
      "command": "node",
      "args": ["/path/to/happypod-mcp/dist/stdio.js"],
      "env": { "HAPPYPOD_API_KEY": "hp_..." }
    }
  }
}

dist/stdio.js is a single bundled file: no npm install needed to run it.

Try it without an account

Set HAPPYPOD_MODE=demo (or pass --demo). Every tool then talks to a built-in simulator: a sample project, deployments that build and go live (or fail with a real-looking error you can fix), logs, metrics and example prices. Nothing leaves your machine and nothing is billed.

HAPPYPOD_MODE=demo node dist/stdio.js

Related MCP server: helipod-mcp

Configuration

Variable

HAPPYPOD_API_KEY

Your API key from happypod.io → Settings → API keys

HAPPYPOD_API_URL

Optional, another API base URL. Default https://api.happypod.io/v1

HAPPYPOD_MODE

demo to use the built-in simulator

Tools

Group

Tools

Account and catalogue

whoami, list_regions, list_sizes, list_gpu_types

Projects

list_projects, create_project, delete_project

Ship

list_pods, get_pod, create_pod, create_gpu_pod, deploy_pod, list_deployments, get_deployment, get_deployment_logs

Operate

get_pod_logs, get_pod_metrics, diagnose_pod, restart_pod, stop_pod, start_pod, scale_pod, rollback_pod, delete_pod

Variables and domains

list_env, set_env, unset_env, add_domain, list_domains

Databases

list_databases, create_database, get_database_connection, delete_database

Every tool that changes something is marked destructive, so MCP clients ask before running it. Paid tools take dryRun: true (returns the price, changes nothing) and an idempotencyKey (a retry never creates twice). Deletes need the resource's exact name in confirm. Errors come back as { status, code, error, details } so the agent can act on them.

Things to ask

  • "Ship this repo to Happy Pod with a Postgres database."

  • "Why is checkout slow in production?"

  • "The last deploy broke the login page, roll it back."

  • "Serve Llama 3.1 8B on the cheapest GPU and stop it when I'm done."

  • "What am I paying for this month, and what can I turn off?"

Develop

npm install
npm test          # tool catalogue, demo flows and the bundled stdio server
npm run build     # bundles src/stdio.ts into dist/stdio.js

The API contract the server expects is in docs/API.md.

License

MIT

Available Tools

33 tools
add_domainAdd custom domainA
Destructive

Attach a custom domain to a pod's public port. Returns the DNS record the user must create (CNAME for subdomains, A/AAAA for an apex). HTTPS is issued automatically once DNS points here.

Inputs: podId (required), hostname (required).

create_databaseCreate databaseA
Destructive

Create a managed Postgres or Redis database with daily backups. The hobby plan is free. dryRun first for paid plans. Then pass its connection URL to the app as a secret (get_database_connection + set_env).

Inputs: name (required), engine (required), projectId (required), plan, +3 more.

create_gpu_podCreate GPU podA
Destructive

Start a GPU pod from a container image (inference servers, notebooks, fine-tuning, batch jobs). Billed per hour while running. ALWAYS dryRun first and tell the user the hourly price; stop it with stop_pod when the job is done.

Inputs: name (required), image (required), gpuType (required), projectId (required), +8 more.

create_podCreate podA
Destructive

Create an app pod and start its first deployment. Source is a Git repository (built automatically: a Dockerfile is used when present, otherwise the stack is detected) or a container image. ALWAYS call with dryRun: true first, tell the user the price, and create only after a clear yes. Then poll get_deployment until success or failed.

Inputs: name (required), projectId (required), env, repo, +10 more.

create_projectCreate projectB
Destructive

Create an empty project (free). One project per app is a good default.

Inputs: name (required), idempotencyKey.

delete_databaseDelete databaseA
Destructive

Delete a database and all its data and backups. Irreversible. Only after the user agrees; confirm must be the database's exact name.

Inputs: confirm (required), databaseId (required).

delete_podDelete podA
Destructive

Delete a pod, its deployments, domains and volumes. Irreversible. Only after the user agrees; confirm must be the pod's exact name.

Inputs: podId (required), confirm (required).

delete_projectDelete projectA
Destructive

Delete an EMPTY project. Irreversible. Refused with 409 PROJECT_NOT_EMPTY while pods or databases remain. Only after the user agrees; confirm must be the project's exact name.

Inputs: confirm (required), projectId (required).

deploy_podDeploy podA
Destructive

Build and roll out the pod again from its source (latest commit of the branch, or the image tag). Needed after changing variables. Zero-downtime: the old version serves until the new one is healthy.

Inputs: podId (required), idempotencyKey.

diagnose_podDiagnose podA
Read-onlyIdempotent

Happy Pod's own health check for a pod: reads the last deployment, recent error logs and metrics, and returns findings (severity, evidence, suggested fix). Start here when the user says something is broken or slow.

Inputs: podId (required).

get_database_connectionDatabase connectionA
Read-onlyIdempotent

Host, port, database and user, plus the connection URL with the password masked. reveal: true returns the real URL: use it only to pass into set_env secrets, never print it to the user.

Inputs: databaseId (required), reveal.

get_deploymentGet deploymentA
Read-onlyIdempotent

One deployment's status: queued, building, deploying, success or failed. Poll every few seconds after create_pod or deploy_pod; builds take one to five minutes.

Inputs: deploymentId (required).

get_deployment_logsDeployment logsA
Read-onlyIdempotent

Build and start-up log of a deployment. Read it when a deployment failed: the cause is usually a missing variable, the wrong port or a build error.

Inputs: deploymentId (required), tail.

get_podGet podA
Read-onlyIdempotent

One pod in full: source, size, replicas, region, ports, URL, domains, status and the last deployment.

Inputs: podId (required).

get_pod_logsPod logsB
Read-onlyIdempotent

Runtime logs of a running pod across replicas.

Inputs: podId (required), tail, level, since.

get_pod_metricsPod metricsB
Read-onlyIdempotent

CPU, memory, requests per second, error rate and p95 latency over a window, per replica and in total.

Inputs: podId (required), window.

list_databasesList databasesC
Read-onlyIdempotent

Managed databases in a project: engine, plan, status, region.

Inputs: projectId (required).

list_deploymentsList deploymentsA
Read-onlyIdempotent

A pod's deployments, newest first: id, status, commit or image, start and finish time. Use it to pick a deploymentId for rollback_pod.

Inputs: podId (required), limit.

list_domainsList domainsB
Read-onlyIdempotent

A pod's domains with DNS and certificate status (pending_dns, issuing, active).

Inputs: podId (required).

list_envList variablesB
Read-onlyIdempotent

A pod's environment variables. Secret values come back masked.

Inputs: podId (required).

list_gpu_typesList GPU typesA
Read-onlyIdempotent

GPU types for GPU pods: VRAM, hourly USD price and the regions with free capacity right now.

Inputs: region.

list_podsList podsB
Read-onlyIdempotent

Pods in a project: app pods (web apps, APIs, workers) and GPU pods, with status, size, replicas, URL and the last deployment.

Inputs: projectId (required).

list_projectsList projectsA
Read-onlyIdempotent

Projects this key can reach, each with its pod and database counts. Start here to find a projectId.

Inputs: no named parameters.

list_regionsList regionsA
Read-onlyIdempotent

Regions where pods and databases can run, with their city and whether GPUs are available there. Pick the one closest to the app's users.

Inputs: no named parameters.

list_sizesList pod sizesA
Read-onlyIdempotent

CPU pod sizes (vCPU, memory, hourly and monthly USD price). Start small: most vibe-coded apps run on small.

Inputs: no named parameters.

restart_podRestart podA
DestructiveIdempotent

Restart the running replicas without rebuilding. Brief interruption per replica.

Inputs: podId (required).

rollback_podRoll back podA
Destructive

Put an earlier successful deployment back live (no rebuild). The quickest fix when a new deploy broke the app.

Inputs: podId (required), deploymentId (required).

scale_podScale podA
DestructiveIdempotent

Change a pod's size and/or replica count. dryRun: true first, show the user the current and new monthly price, apply after a clear yes.

Inputs: podId (required), size, dryRun, replicas.

set_envSet variablesA
DestructiveIdempotent

Create or update variables. Put credentials in secrets. Changes apply on the next deploy: call deploy_pod afterwards.

Inputs: podId (required), env, secrets.

start_podStart podA
DestructiveIdempotent

Start a stopped pod again. Billing resumes.

Inputs: podId (required).

stop_podStop podA
DestructiveIdempotent

Stop a pod (billing for compute stops; volumes are kept). GPU pods should be stopped as soon as their job is done.

Inputs: podId (required).

unset_envRemove variableB
DestructiveIdempotent

Remove one variable. Applies on the next deploy.

Inputs: key (required), podId (required).

whoamiWho am IA
Read-onlyIdempotent

The account behind the API key: name, plan, credit balance (USD), spend this month and the key's scope. Call it first in a session. mode "demo" means the built-in simulator, not a real account.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 33 tools

    Disambiguation4/5

    Most tools map cleanly to a unique resource and action, with strong descriptions explaining adjacent pod lifecycle operations. A few could be confused at a glance—start_pod/restart_pod/deploy_pod/rollback_pod and create_pod/create_gpu_pod—but the descriptions usually disambiguate.

    Naming Consistency5/5

    The set follows a very consistent snake_case verb_noun pattern: list_pods, get_pod, create_database, delete_project, set_env, etc. whoami is the only exception, but it is a conventional standalone command and does not undermine the pattern.

    Tool Count3/5

    33 tools is heavy for an agent-facing MCP surface, even for a broad PaaS. Most tools are genuinely distinct, but several pod lifecycle actions could potentially be consolidated, making this borderline rather than clearly excessive.

    Completeness4/5

    The core deployment lifecycle is well covered: projects, pods, GPU pods, env vars, domains, databases, deployments, logs, metrics, and diagnostics. Notable gaps remain around domain deletion, database resize/backup-restore, and deployment cancellation, but core workflows are usable.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI assistants to create projects, deploy files, container images, or Git repositories, attach domains, set write-only secrets, add Postgres, read logs, metrics, and failure summaries, and roll back.
      19 npm
      Apache 2.0
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI coding tools to deploy apps and databases, run VPS servers, buy domains and edit DNS, and send email through natural-language requests across Container, VPS, Domain, and Email APIs.
      350 npm
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI assistants to deploy and manage hosted apps from git repos, read logs, add Postgres databases and domains, manage API keys and budgets, and send requests to 200+ AI models. Connects over remote Streamable HTTP with OAuth sign-in, so users need no API keys, installs, or local server.
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to build and deploy an app from a GitHub repository to a managed live server and then operate it — logs, captured errors, health checks, managed Postgres, secrets, custom domains, email authentication, and Stripe checkout links. It also exposes safe, read/write-labelled tools so agents can rebuild, start/stop, and inspect projects without ever editing source code.
      MIT