temporal-python-testing skill for Claude Code: test Temporal workflows
Teaches Claude to write and run pytest tests for Temporal workflows in Python, for backend developers who build on Temporal.
- Installs
- 11.1K
- on skills.sh
- Stars
- 42.1K
- on GitHub
- Files
- 4
- besides SKILL.md
- Updated
- Oct 2026
- last repo commit
The skill is a testing guide that Claude loads when you write Temporal workflow tests or debug failing ones. It covers unit tests with time-skipping, integration tests with mocked activities, replay tests that check determinism against production histories, and a local setup with Docker Compose. A short quick start shows a workflow test and an activity test.
SKILL.md keeps the overview and points Claude to four resource files, which it loads for the scenario you ask about. It bundles no scripts. It leans on the temporalio, pytest and pytest-asyncio packages, and the local setup guide uses Docker.
Good for
- Python developers writing Temporal workflows and activities
- Teams adding replay tests to CI before deploying workflow changes
- Anyone setting up a local Temporal server for tests
Install temporal-python-testing
Claude Code
- 1
Copy this prompt.
Install the temporal-python-testing skill in Claude Code. Run this command: npx skills add wshobson/agents --skill temporal-python-testing Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js.
Pick the skills CLI if you only want this skill. The plugin also installs 8 agents, such as temporal-python-pro and test-automator, and 1 command. Your project needs Python and the temporalio package.
Built from the package data we hold. Not yet tested by us. Source.
What's in the skill
When it loads
Test Temporal workflows with pytest, time-skipping, and mocking strategies. Covers unit testing, integration testing, replay testing, and local development setup. Use when implementing Temporal workflow tests or debugging test failures.
Sections in SKILL.md
- When to Use This Skill
- Testing Philosophy
- Available Resources
- Unit Testing Resources
- Integration Testing Resources
- Replay Testing Resources
- Local Development Resources
- Quick Start Guide
- Basic Workflow Test
- Basic Activity Test
- Coverage Targets
- Key Testing Principles
- How to Use Resources
- Additional References
Key facts
- Author
- Seth Hobson (wshobson)
- Repository
- wshobson/agents
- Installs on skills.sh
- 11,133 as of 2026-11-02
- GitHub stars
- 42,081 for the whole wshobson/agents repo
- Last repo commit
- 2026-10-30
What it includes
Time-skipping workflow tests and activity tests in isolation
Covers WorkflowEnvironment with time-skipping, ActivityEnvironment, manual time advancement and pytest fixtures.
Try: “Show me unit testing patterns for my Temporal workflow.”
Read ·
resources/unit-testing.mdMocked activities, error injection, signals and queries
Covers activity mocking, retry and non-retryable error tests, multi-activity workflows and coverage strategies.
Try: “How do I mock activities in this workflow test?”
Read ·
resources/integration-testing.mdCheck determinism against recorded workflow histories
Shows how to export histories from a Temporal server, replay them with current code and run that in CI. The export examples connect to a Temporal server address.
Try: “Validate that my workflow change is deterministic.”
Network ·
resources/replay-testing.mdDocker Compose server, pytest config, coverage and a Makefile
The guide has you run docker-compose, pip install the test packages and set up pre-commit and Makefile tasks. Its clean target removes the Docker volume and test caches.
Try: “Set up a local Temporal server for my tests.”
Runs code ·
resources/local-setup.mdA basic workflow test and a basic activity test
The workflow test uses WorkflowEnvironment.start_time_skipping() with a Worker. The activity test uses ActivityEnvironment.
Quick start testsAim for at least 80% logic coverage on workflows and activities
It also asks for replay tests of all workflow versions before deployment.
Coverage targets
From the official docs
Permissions
Low risk
The skill is guidance, but its guides tell Claude to run Docker, install packages and export workflow histories. Run those steps on a dev machine and check what the histories contain before you commit them.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
Its setup guide starts containers and installs packages
The Docker Compose example runs a Temporal server and a Postgres database with the password temporal. Use it on a dev machine only, and change the password if the ports are reachable from outside. github.com
Its Makefile example deletes data and caches
The clean target runs docker-compose down -v, which removes the Postgres volume, and rm -rf on .pytest_cache, htmlcov and .coverage. Check the target before you run it. github.com
History export reads from a Temporal server
The replay guide shows scripts that connect to a server address and save up to 100 workflow histories to files. Those histories can hold workflow inputs, so keep the files out of public repos. github.com
Technical details
| Turning on read-only | The skill defines no read-only switch. It bundles no scripts. |
|---|---|
| Talks to | Docker image pulls for the Compose setup (temporalio/auto-setup, postgres, temporalio/ui). Your Temporal server address, in the history export examples. pip package downloads from the install commands |
| Scopes |
|
Questions
What does the temporal-python-testing skill do?
It gives Claude patterns for testing Temporal workflows in Python. It covers unit tests with time-skipping, mocked-activity integration tests, replay tests for determinism, and a local server setup.
How do I install temporal-python-testing?
Use the install box at the top of this page. The skills CLI installs just this skill. The backend-development plugin installs it with 8 agents and 1 command.
Does it need an API key or a Temporal account?
The SKILL.md names no API key. Its local setup guide runs Temporal with Docker Compose on your machine, and its replay examples connect to a Temporal server address you provide.
Is it safe to install?
It is guidance files and bundles no scripts. The guides ask Claude to run docker-compose, pip install and a Makefile that can delete a Docker volume, so run those on a dev machine and read the commands first.
Does it work in the Claude app or Codex?
The install box has a Claude app upload and a Codex route. The Claude app can write the tests, but you run pytest and the Temporal server on your own machine.
Why does it want 80% coverage?
The SKILL.md recommends at least 80% logic coverage for workflows and activities and cites docs.temporal.io best practices for it. It is a target in the guide, not a check the skill runs.
How is it different from temporal-python-pro?
temporal-python-pro is an agent in the same backend-development plugin, described as covering workflow design, saga patterns and testing strategies. This skill is only the testing guide, and you can install it without the agents.
Information
updated Nov 2, 2026
- Repository
- wshobson/agents
- Folder
plugins/backend-development/skills/temporal-python-testing
Sources
Docs, issue trackers and discussions this page is built from.
Show all 8Hide
- wshobson/agents README · github.com
- wshobson/agents marketplace.json · github.com
- temporal-python-testing SKILL.md · github.com
- local-setup.md · github.com
- unit-testing.md · github.com
- integration-testing.md · github.com
- replay-testing.md · github.com
- temporal-python-pro agent · github.com