InnerVault

Kubernetes Ephemeral Job

Stale Official MCP Registry, io.github.inhuman

Runs a command in a throwaway Kubernetes pod and returns exit code, output and artifacts.

Open the source (opens github.com) https://github.com/inhuman/mcp-k8s-ephemeral-job

Before you use it

  • Runs on your computer with your permissions. Read what it can do before connecting it to an agent.
  • Needs credentials: MCP_K8S_AUTH_TOKEN, MCP_K8S_CLONE_SECRET. Use a key with the least access you can.

What it is

MCP servers from io.github.inhuman.

The problem it solves and who it suits haven't been reviewed by InnerVault. The description above is the source's own.

How to get it

oci package (stdio)

docker.io/idconstruct/mcp-k8s-ephemeral-job:v0.7.1

From the server's entry in the official MCP Registry

Settings it asks for

  • MCP_K8S_ALLOWED_IMAGES required · Strict image allowlist (CSV). Empty means every call is rejected, so this must be set.
  • MCP_K8S_KUBECONFIG · Path to the kubeconfig inside the container. Empty = use in-cluster credentials.
  • MCP_K8S_NAMESPACE · Namespace the ephemeral pods are created in. Needs a Role/RoleBinding for jobs and pods.
  • MCP_K8S_TRANSPORT · MCP transport: stdio | http | sse. Must be 'stdio' for direct docker/stdio use.
  • MCP_K8S_DEFAULT_TIMEOUT_S · Default wall-clock timeout per job, seconds.
  • MCP_K8S_MAX_TIMEOUT_S · Maximum wall-clock timeout a caller may request, seconds.
  • MCP_K8S_MAX_OUTPUT_BYTES · Cap on combined stdout/stderr before truncation.
  • MCP_K8S_MAX_ARTIFACT_BYTES · Cap on the total size of returned artifacts.
  • MCP_K8S_MAX_CONCURRENT · Maximum ephemeral pods running at the same time.
  • MCP_K8S_DEFAULT_CPU · Pod CPU request; limits come from the caller or the namespace LimitRange.
  • MCP_K8S_DEFAULT_MEMORY · Pod memory request; limits come from the caller or the namespace LimitRange.
  • MCP_K8S_SIDECAR_IMAGE · Helper image used to inject input files and collect artifacts.

Related

  1. no-crd

    MCP servers

    Dynamic pod spawner & proxy for ephemeral AI agent workspaces on Kubernetes without CRDs

    Stale Official MCP Registry, io.github.nogoo9 Updated 3 months ago 2 notes

    @nogoo9/no-crd@0.18.0
  2. cronstable

    MCP servers

    A job scheduler with retries, alerts, saved run history, workflows, and dashboards for single machines and clusters.

    Stale GitHub, ptweezy 138 ★MITUpdated 4 days ago

    git clone https://github.com/ptweezy/cronstable.git
  3. Ephemeral tool forge

    MCP servers

    Ephemeral single-task tool forged from examples, sandbox-proven, task-bound, expiring by design.

    Stale Official MCP Registry, com.textforagents Updated 20 days ago 1 note

  4. radar

    MCP servers

    The missing open-source Kubernetes UI with a built-in MCP server for AI agents. See what's broken, why, and what changed. Issues, Topology, event timeline, Helm, GitOps, live service traffic, and cluster audits - all in one Go binary.

    Indexed GitHub, skyhook-io 4,131 ★Apache-2.0Updated today

    git clone https://github.com/skyhook-io/radar.git
  5. Ephemeral

    MCP servers

    Run AI agent tasks in isolated sandboxes with a cost cap, saved workflows and scoped connections.

    Stale Official MCP Registry, al.ephemer Updated 21 days ago 1 note

  6. ephemeral-cloud-mcp

    MCP servers

    Temporary webhooks, URL watches and delayed callbacks for AI agents. Pay per use in USDC (x402).

    Stale Official MCP Registry, io.github.weekendinvestor Updated 56 days ago 1 note

    ephemeral-cloud-mcp@0.1.3