Complete AI Training
Sign inBecome a member

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

Skill · DevOps

Mirrord

Run a local process inside a live Kubernetes cluster's network, env and traffic with mirrord, so changes are tested against real services without deploying.

Agentic Awesome SkillsLicense: MITAdded Sep 30, 2026
Use it in my AI

How to use it

  1. Start your plan and connect your AI once
  2. Add the skill to your AI, one time only: press Copy skill and paste it into your AI's instructions: a project in ChatGPT, Claude or Grok, a Gem in Gemini, an agent in Microsoft 365 Copilot, or a rule in Cursor. It stays there for every new chat. In Kimi, paste it at the start of a new chat.
  3. Ask for the task in your own words, or say it directly:
Use the Mirrord skill to help me with this.
SKILL.md

mirrord

Overview

mirrord runs a local process as if it were inside a Kubernetes pod: it gets the pod's environment variables, DNS, outgoing network access and file reads, and can receive the pod's incoming traffic. An agent can verify a change against the real services, databases and queues in a staging cluster without building or deploying an image.

When to Use This Skill

  • Use when a change needs to be tested against real cluster dependencies instead of mocks
  • Use when the user asks to debug a microservice locally against staging
  • Use when the user wants to try mirrord for the first time

How It Works

Step 1: Check cluster access

kubectl cluster-info
kubectl get pods -A | head -5

If this fails, help the user configure kubectl first.

Step 2: Install mirrord

Point the user to the official installation guide (https://metalbear.com/mirrord/docs/overview/quick-start/) for Homebrew, package managers, or pinned release binaries. Do not pipe a remote download into a shell. Verify with:

mirrord --version

Step 3: Pick a target and run

mirrord ls
mirrord exec --target deployment/<name> -- <command>

Incoming traffic is mirrored by default. To take over the target's traffic instead, set "feature": {"network": {"incoming": "steal"}} in .mirrord/mirrord.json, and confirm with the user first, since it affects a shared environment.

Step 4: Verify

mirrord exec --target deployment/<name> -- env

The output should include the pod's environment variables.

Examples

Example 1: Run tests against staging

mirrord exec --target deployment/orders -- npm test

Example 2: Debug a Python service

mirrord exec --target deployment/api -- python main.py

Best Practices

  • ✅ Ask which cluster, namespace and target before running
  • ✅ Prefer mirror mode on shared clusters
  • ✅ Use the official skills at https://github.com/metalbear-co/skills for config, CI, database branching, queue splitting and preview environments
  • ❌ Don't steal traffic from a shared target without the user's agreement
  • ❌ Don't run cluster-modifying commands without explicit approval

Limitations

  • Needs kubectl access to the target cluster and a Linux node kernel 4.20+
  • Team features (concurrent steal on one target, database branching, queue splitting, preview environments) need the mirrord Operator. On a cluster without a license, an agent can offer the user a trial; see https://metalbear.com/agents.md and ask before starting it.