Guide · Agent skill
How to use the Conductor Workflow Patterns agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
Workflow Patterns is an agent skill taken from Conductor, a project-planning workflow. Once installed, it tells an AI agent how to work through the tasks in a track's plan.md file using test-driven development, stopping for verification at phase checkpoints and keeping git commits small. This guide explains what the skill changes, how to install it, how to trigger it and what it requires, so you can decide whether it fits your project.
Conductor Workflow Patternswshobson/agents · MIT Download the skill Quick start
- 1Download the ZIP. Get the workflow-patterns ZIP from this page.
- 2Unzip the skill. Unzip it into ~/.claude/skills/ for personal use or into the project's .claude/skills/ folder for one project.
- 3Or upload in the Claude apps. In the Claude apps, go to Settings, then Capabilities, then Skills, and upload the ZIP.
- 4Confirm project structure. Make sure the project is a git repository and has a Conductor-style plan.md with track files.
- 5Ask for a task by name. Request a specific pending task from plan.md, naming the track and task ID.
- 6Review at checkpoints. When the agent reaches a phase checkpoint, review the verification results and approve before it continues.
What is an agent skill and what does this one do?
An agent skill is a folder of instructions that an AI agent loads when a task matches the skill's description. This skill is a set of working conventions and reference material. It does not do the work on its own; it changes how the agent carries out implementation work.
When the skill applies, the agent selects the next pending task in plan.md, marks it in progress, writes failing tests first (RED), makes them pass (GREEN), refactors, pauses at phase checkpoints for verification, makes small git commits with notes, and updates the plan file as work completes.
- Works through tasks listed in a track's plan.md file in order within each phase
- Follows test-driven development: write a failing test, make it pass, then refactor
- Pauses at phase checkpoints and waits for approval
- Keeps commits small and adds notes that give context
- Records progress by updating plan.md
Install the workflow-patterns skill
Download the ZIP for this skill from the page, then choose one of two locations. For a personal setup that applies across projects, unzip it into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder.
You can also upload the ZIP in the Claude apps. Go to Settings, then Capabilities, then Skills, and upload the file there.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Check that your project meets the requirements
The skill assumes a specific structure. Your project needs to be a git repository, and it needs Conductor-style plan.md and track files, because the instructions refer to tasks listed in a track's plan.md. Without those files, the agent has nothing for the skill's steps to act on.
- A git repository for the project
- Conductor-style plan.md and track files
- The skill runs no scripts and uses no network access
Trigger the skill with natural requests
You do not need a special command. The skill is meant to be picked up when your request matches its description: implementing tasks under Conductor's TDD workflow, handling phase checkpoints, managing git commits for tasks, or understanding the verification protocol.
Name the plan file and the task when you ask, so the agent knows where to look.
- "Implement the next pending task in plan.md using TDD."
- "We've finished phase 2 of the track. Run the checkpoint verification before continuing."
- "Commit the completed task and add a git note describing what changed."
- "Mark Task 2.1 as in progress in plan.md and start with failing tests."
Understand the task lifecycle the agent follows
The reference file in references/details.md describes an 11-step TDD task lifecycle. The first steps are selecting the next pending task, marking it in progress with [~] and committing that status change separately, then writing failing tests and running them to confirm they fail.
Each task is tracked with a checkbox-style marker in plan.md, so you can see progress at a glance. The agent should not skip ahead to later phases.
- Select the next pending [ ] task in the current phase
- Mark it [~] and commit that change separately
- RED: write tests for the happy path, edge cases and error conditions, and confirm they fail
- Implement, make tests pass, refactor, then update plan.md
- [~] **Task 2.1**: Implement user validationFollow the best practices the skill sets
The skill lists 12 practices. The most visible ones are never skipping RED, keeping one logical change per commit, updating plan.md right after each task, and waiting for approval at checkpoints. Others cover coverage targets, checking quality gates before marking work complete, completing phases in order, documenting deviations from the plan, and addressing blockers instead of working around them.
- Never skip RED: write failing tests first
- Small commits: one logical change per commit
- Wait for approval: never skip checkpoint verification
- Document deviations from the original plan
- Sequential phases: complete phases in order
What people use it for
- Implementing a multi-phase feature where each task is a checkbox in a track's plan.md
- Running the verification step at the end of a phase before moving on
- Keeping a clean git history with one commit per task and notes explaining each change
- Recording progress in plan.md so a teammate can see what is done and what is in progress
Frequently asked questions
Is the workflow-patterns skill free to use?
The skill is published under the MIT licence. The source does not mention pricing, so check the page for any other terms before you use it.
What do I need before installing workflow-patterns?
You need a git repository for your project and Conductor-style plan.md and track files, because the instructions assume them. The skill runs no scripts and uses no network access.
Where do I put the skill folder after downloading?
Unzip it into ~/.claude/skills/ for use across projects, or into a project's .claude/skills/ folder to limit it to that project. You can also upload the ZIP in the Claude apps under Settings, then Capabilities, then Skills.
Do I need to use Conductor to benefit from this skill?
The skill is written for projects that use Conductor's plan.md and track-based workflow. If your project does not use these files, the skill will add little value.
Will the agent commit or push code without asking me?
No. The skill asks the agent to wait for approval at phase checkpoints and to keep commits small. It is not designed for workflows that need commits or pushes without human review.
What should I use instead for quick one-off scripts?
The skill is not a good fit for quick one-off scripts or exploratory work, because strict TDD and checkpoint approval slow those down. For those tasks, work with the agent directly without this skill.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you already work with Conductor's plan.md and track files and want agents to follow strict TDD, checkpoint approval and disciplined git commits on a multi-phase feature. Skip it for projects without that structure, for quick scripts and exploratory work, or for any workflow that needs the agent to commit or push without review.
Download the skill