A plan you can build against
Turn an idea into a plan with acceptance criteria. Work locally, or connect the Zensu CLI to bring in a tracked feature's spec, journeys and security context.
Zensu for GitHub Copilot CLIFrom a plan to code
From a plan to code
you can verify.
Give your Copilot session a complete development workflow: plan, implement, verify and review. Zensu runs natively in Copilot CLI, keeping the evidence with the work from the first change to the final review.
Private marketplace · GitHub repository access required
> /zensu-tdd Build a minimal browser Pong plan scope · acceptance criteria implement & verify code · tests · browser evidence review & resolve native reviewers → fixes → self-review finish current evidence · completed review chain
GitHub Copilot CLI in about three minutes
From an idea to a reviewed pull request.
See the Zensu workflow in GitHub Copilot CLI, from an agreed plan to verified code. The film explains Zensu Autopilot and how it differs from Copilot's own execution mode.
A clear starting point
Turn a request into a plan with acceptance criteria you can check.
Evidence along the way
Review the change, fix concrete findings and verify the running feature.
The final decision is yours
Zensu Autopilot prepares a pull request. You control review, merge and deployment.
Read the video transcript
A clear path from request to reviewed code
Writing code is only one part of shipping a feature. You also need a clear plan, meaningful checks, and a review you can act on. Zensu brings those steps into GitHub Copilot CLI, so an idea can move through a consistent delivery workflow.
On screen: A feature moves through planning, building, review, verification and a pull request.
One real-world task
Imagine your team needs a CSV export for filtered orders. You describe the outcome in plain English: export the orders currently shown, include the right columns, and respect access permissions. Zensu helps turn that request into a plan with concrete acceptance criteria.
On screen: A fictional Orders app filters a date range. The plan covers the date filter, expected columns and access permissions.
Agree on the outcome
Before implementation begins, you review the plan. For an end-to-end run, choose Zensu Autopilot. This is Zensu's feature-to-pull-request workflow, separate from Copilot's own Autopilot execution mode. After that planning gate, Zensu works through building, reviewing, fixing, and verifying the feature, aiming for a pull request that is ready for your review.
On screen: You approve the plan. Zensu Autopilot moves through building, review, fixes and verification toward a pull request. Zensu Autopilot is explicitly distinguished from Copilot's own Autopilot execution mode.
Build with the right level of guidance
During implementation, your coding agent works in your repository and follows its existing conventions. Zensu uses a flexible implementation mode by default. If your project calls for strict test-driven development, you can opt into a failing test-first, red-to-green cycle.
On screen: Implementation follows the existing repository. Strict test-driven development is an optional failing-test-first, red-to-green cycle.
Bring in fresh perspectives
Specialist reviewers examine the change from different angles: correctness, design, and security. Findings must point to actual files and lines, and are checked against the source before they drive fixes. An independent judge checks for blind spots.
On screen: Review findings identify source files and lines. An independent judge checks for blind spots.
Make feedback actionable
In our example, a reviewer spots that the export ignores the selected date filter. That identifies a specific gap between the requested behavior and the implementation. The workflow applies a fix and reviews the resulting change.
On screen: The export initially ignores the date filter. The fix passes the filter to the order lookup, then the change is reviewed again.
Check the running feature
Live verification exercises the running feature against the acceptance criteria. Select a date range, export the orders, and inspect the downloaded file. Do the rows match the filter? Are the columns correct? Are access permissions enforced? The report records what actually happened, including anything that remains unresolved.
On screen: The downloaded CSV contains the expected rows and columns. An unauthorized request receives 403. An acceptance report records the outcomes.
Keep the final decision
When the checks pass, the result is a pull request with implementation and verification evidence. Autopilot stops there. You retain control over the final review, merge, and deployment.
On screen: The pull request carries implementation, review and verification evidence. Final review, merge and deployment remain your decisions.
Connect work to product progress
Zensu also supports the bigger picture. Starting fresh? Bootstrap turns a vision document into tracked features. Working with an existing product? Ghost Scan discovers what is already built. With the optional Zensu tracking layer, features connect to source files, tests, documentation, journeys, security profiles, and a dashboard your team can follow.
On screen: Bootstrap starts from a vision; Ghost Scan discovers existing features. Optional tracking connects features to their evidence.
Start with one change
You do not need a Zensu account for the core implementation workflow and review chain. Start with one change, or connect product tracking when you need it. Less manual coordination. Clearer evidence. A more understandable path from request to reviewed code.
On screen: The core workflow needs no Zensu account. Product tracking is optional. Start with /zensu-setup and visit zensu.dev/copilot.
From idea to reviewed change
A workflow that stays with the feature.
Start with a local project. Connect your product record when you need shared feature tracking. Local implementation and review can work without a Zensu account.
Review before you call it done
Native Copilot agents review the change. Findings enter a review and fix cycle, followed by a final self-review. Code, tests and evidence stay together.
Completion backed by evidence
An active Zensu workflow checks implementation evidence and review completion. Missing or stale evidence keeps it open, so the next session can see what remains.
Your first session
Approve the plan. Choose the workflow.
Copilot's plan approval and Zensu's workflow choice are separate steps. To make that choice interactively, use this sequence.
Plan in Copilot
Use Copilot's Plan mode to develop the scope and acceptance criteria. It presents its own “Plan ready for review” dialog.
Continue interactively
Choose “Accept plan and continue in Interactive execution mode” when you are ready to implement.
Make the Zensu choice
With the default ask setting and no prior choice for the session, Zensu asks whether to use its workflow. You can choose Zensu or direct implementation.
Two different meanings of Autopilot
Copilot's built-in Autopilot controls autonomous continuation and can skip the interactive workflow question. /zensu-autopilot is Zensu's feature-to-PR workflow. Choose Interactive execution mode when you want to answer the workflow question yourself.
Built for Copilot CLI
Native tools. The full Zensu workflow.
32 skills and six native review and product-lifecycle agents cover planning, implementation, security, documentation, review and recovery. The Zensu CLI connects to your product record; MCP is not a prerequisite.
Vanilla by default. Test-first by choice.
The default workflow lets you implement directly, then verify and complete the review chain. Enable strict RED → GREEN with /zensu-tdd-mode when you want failing-test evidence before production edits.
Copilot sessions and permissions
Zensu uses Copilot's own tools, agents, session identity and approval prompts. Browser validation uses playwright-cli with permission checks. Claude Code is not required.
Install and update
Install through Copilot's native marketplace.
The Zensu repository and marketplace are private. Your GitHub account needs read access to MKITConsulting/zensu-copilot. Request repository access via [email protected] before installing.
1. Prepare your CLI
Use an authenticated GitHub Copilot CLI, Node.js 22 or newer, Bash and Git. This plugin targets Copilot CLI; the VS Code extension and GitHub's hosted coding agent are outside this guide.
For tracked features, Pulse and work orders, install the Zensu CLI and run zensu auth login. GitHub PR workflows need authenticated gh; GitLab MR workflows need glab.
2. Add the marketplace
Run these commands in your terminal. Keep the #marketplace suffix: it selects the distribution branch.
copilot plugin marketplace add MKITConsulting/zensu-copilot#marketplace
copilot plugin install zensu@zensuThe release catalogue pins the plugin to an immutable commit. No local build is needed.
3. Start a fresh Copilot session
Restart Copilot, then run these commands inside the conversation to discover the skills, check prerequisites and configure your project:
/zensu-help
/zensu-doctor
/zensu-setup --projectFor browser verification, Doctor reports the required Playwright CLI tooling. Other validation drivers may need additional tools.
Keep the plugin up to date
New releases are published to the marketplace. Refresh its catalogue, update the plugin, then restart Copilot:
copilot plugin marketplace update zensu
copilot plugin update zensu@zensuThe 0.2.1 release has been checked live on macOS, including a complete workflow, browser navigation and interactive approval and rejection.
Already installed Zensu directly from a local directory?
Run copilot plugin list --json to identify the old registration. Remove that registration with copilot plugin uninstall <direct-install-name> before installing zensu@zensu. Keeping both can load duplicate hooks. Preserve your Zensu configuration and project state, then start a fresh session.
Integration support
Bring Zensu into your next Copilot session.
For repository access or help with setup, contact [email protected]. To connect your workflow to the Zensu product platform, request an account below.
Request Zensu access