Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Docker Compose Local Dev Environment Builder and Audit

    1

    Detects your stack and writes compose.yaml with healthchecks, Compose Watch hot reload, .env.example, profiles and make shortcuts. Verdict READY / NEEDS WORK / BROKEN

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Docker Compose Local Dev Environment Builder and Audit

    Docker Compose Local Dev Environment Builder and Audit

    Example session with this skill installed

    Audit the docker-compose.yml in my Django repo and turn it into a proper one-command dev setup. We use Postgres, Redis and Celery, half the team is on M-series Macs, and the web container keeps crashing on startup because the DB isn't ready.

    • Read your context and instructions
    • Compiled the docker compose local

    Stack detected

    ComponentEvidenceChoice
    Django 5 apppyproject.toml, manage.pybuild target: dev, port 8000
    PostgreSQLsettings.py:88 uses DATABASE_URLpostgres:16.x (pin exact minor)
    RedisCELERY_BROKER_URL=redis:// in .envredis:7.x
    Celery workercelery.pysame build as app
    MailEMAIL_HOST in settingsMailpit, profile mail

    Verdict: NEEDS WORK

    It starts, but web often boots before Postgres is ready, and data and ports are handled loosely.

    #Issue in current docker-compose.ymlFixSeverity
    1version: "3.8" at topRemove; it is obsolete and only warnsLow
    2depends_on: [db] with no healthcheckpg_isready healthcheck + condition: service_healthyHigh
    3image: postgres:latestPin a 16.x tagMedium
    45432:5432 open on all interfaces127.0.0.1:${DB_PORT:-5432}:5432Medium
    5Real-looking SECRET_KEY in the fileMove to .env, commit .env.example with dummies, rotate the keyHigh
    6Whole repo bind-mounted, slow on macOSdevelop.watch sync of ./app, rebuild on uv.lockMedium

    compose.yaml (excerpt)

    name: shop
    services:
      db:
        image: postgres:16.4
        env_file: [{ path: .env, required: false }]
        volumes: [pgdata:/var/lib/postgresql/data]
        ports: ["127.0.0.1:${DB_PORT:-5432}:5432"]
        healthcheck:
          test: ["CMD-SHELL", "pg_isready -U $$POSTGRES_USER -d $$POSTGRES_DB"]
          interval: 5s
          retries: 10
      migrate:
        build: { context: ., target: dev }
        command: python manage.py migrate --noinput
        depends_on: { db: { condition: service_healthy } }
      web:
        build: { context: ., target: dev }
        ports: ["127.0.0.1:8000:8000"]
        depends_on:
          db: { condition: service_healthy }
          migrate: { condition: service_completed_successfully }
        develop:
          watch:
            - { action: sync, path: ./app, target: /code/app }
            - { action: rebuild, path: uv.lock }
    volumes: { pgdata: {} }
    

    Validation

    • docker compose config --quiet: exit 0
    • docker compose up --wait: not run here (run make up and paste the output)

    Next

    make up, then make watch. Clean DB: make reset (runs docker compose down -v, which deletes all local data).

    compose.yaml

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Most local dev setups grow by copy-paste: a docker-compose.yml with an old version: line, an app that starts before the database is ready, latest tags that change under you, ports open to the whole network, and a bind mount that makes a Mac crawl. Asking an AI to "write a compose file" usually gives a generic template that ignores what your repo actually uses.

    This skill gives your coding agent a fixed procedure and a reference file of facts checked against docs.docker.com. It:

    • Detects the app language, framework, dev command and port, plus databases, caches, queues, search, object storage and mail, citing the file each finding came from
    • Writes compose.yaml (no obsolete version key) with healthchecks, depends_on condition service_healthy, named volumes, pinned tags and ports bound to 127.0.0.1
    • Sets up hot reload with Compose Watch (sync and rebuild rules) or a bind mount fallback
    • Adds .env.example with dummy values, profiles for optional services, a migrate/seed one-off job and Makefile or justfile shortcuts
    • Validates with docker compose config and docker compose up --wait, and reports only results it actually saw
    • Adds platform notes for Apple Silicon, macOS file sharing, Windows and WSL2, plus a troubleshooting table
    • Gives a verdict for an existing setup: READY, NEEDS WORK or BROKEN

    Who it is for: developers and team leads who want new hires to run one command and get a working stack, and anyone whose current compose file is flaky or slow.

    Honesty note: the skill does not invent image tags or Compose features. Unconfirmed tags are marked TODO, and commands it could not run are marked "not run". Data reset is described only as docker compose down -v, with a warning that it deletes local data.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.

    Built for local development only, not production, Swarm or Kubernetes. Image tags and docker commands are verified only when the agent has network access and Docker installed; otherwise they are marked TODO or not run. Healthcheck commands depend on what each pinned image ships, so some may need adjusting.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 29 days ago

    Software
    Docker Desktop
    orDocker Engine
    Docker Compose plugin

    Compose Watch needs Compose v2.22+.

    Permissions

    Network Access
    Write Files
    Listed29 days ago

    What's inside

    Frequently Asked Questions

    A SKILL.md with the step-by-step procedure, verdict rules, output template and checklist, and references/compose-dev-facts.md with Compose facts, Compose Watch actions and minimum versions, healthcheck commands for common services, platform notes, troubleshooting and source links.

    Any agent that reads SKILL.md skills, such as Claude Code, Codex CLI, Cursor, VS Code Copilot and Gemini CLI. Behavior can vary slightly between agents.

    Unzip the docker-compose-dev-environment folder into your agent's skills directory, then ask something like 'set up docker compose for this repo' from the repo root. Docker with Compose v2 is needed to run the validation steps.

    It targets local development, not production or Kubernetes. It cannot confirm image tags or run docker commands without network access and Docker; in that case it marks them TODO or not run. Unusual stacks may need manual tweaks.

    The agent follows a fixed detection and validation procedure, works from a checked fact file instead of memory, cites evidence for each detected service, and returns a consistent report with a READY / NEEDS WORK / BROKEN verdict instead of a generic template.

    No. This skill is built for local development only. For production Kubernetes manifests or Helm charts, see the companion Kubernetes Manifest and Helm Chart Writer and Reviewer skill from the same creator.

    New in DevOps & Cloud