Skip to content
Get packs

DeerFlow Smoke Test

Walks an agent through pulling DeerFlow, deploying it locally or in Docker, and writing a health report

Source Download skill
Repository
bytedance/deer-flow
Stars
84,876
Licence
MIT
Files
12 files

What it does

When installed, this skill gives the agent a step-by-step checklist for checking that the DeerFlow project works end to end. The agent updates the code from git, checks the machine has the right tools, starts the app either locally or in Docker, then checks that the web page, API and key routes respond. It finishes by writing a test report from a template. Scripts are provided for each step so the agent can run the checks rather than guess.

Great for

  • Developers who have just cloned or updated DeerFlow and want to confirm a fresh install runs
  • Maintainers who want a repeatable pre-release or post-deployment verification
  • Teams testing DeerFlow on a new machine or server

Skip it if

  • Anyone who wants to use DeerFlow as an end user without installing it; the skill is for testing a deployment, not running research tasks
  • Machines where you do not want git pulls, package installs or services started automatically, since the workflow changes the project directory and launches processes
  • Production servers with live data, because the workflow runs make targets that start and restart services and may modify config.yaml and .env files

Try asking

  • “Run the smoke test on this DeerFlow checkout and tell me whether it deploys locally”
  • “Verify my Docker installation of DeerFlow and write the test report”
  • “Check whether the frontend routes under /workspace load after deployment”
  • “Run an end-to-end test of the service availability and health endpoints”
Step-by-step guide: how to use DeerFlow Smoke TestInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 12 original files, the MIT licence and an attribution note.

    smoke-test.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip smoke-test.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip smoke-test.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Read the scripts before running them and run the check scripts first, since deploy scripts start services and may create config files
  • Use a disposable clone or test machine rather than a production or heavily customised checkout, because the workflow runs git pull and make targets
  • Keep API keys out of chat messages; let the agent read them from the local .env file and confirm the agent does not send them anywhere

What's inside

  • references/SOP.md
  • references/troubleshooting.md
  • scripts/check_docker.sh
  • scripts/check_local_env.sh
  • scripts/deploy_docker.sh
  • scripts/deploy_local.sh
  • scripts/frontend_check.sh
  • scripts/health_check.sh
  • scripts/pull_code.sh
  • SKILL.md
  • templates/report.docker.template.md
  • templates/report.local.template.md

Redistributed unmodified from bytedance/deer-flow under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.