BCQuality — Don’t teach one agent. Teach the ecosystem. 🤝

Quality skills and knowledge that help AI tools make better Business Central development decisions: catch BC-specific defects, avoid misleading advice, and explain findings with references you can read.

BCQuality contains knowledge and reusable skills, not agents or a Business Central extension. Your host supplies the agent. You can install the content as a plugin, use it from another integration, or browse the knowledge directly.

Three ways to use BCQuality

Use	Status	Start here
Review your AL code on demand with the Copilot CLI plugin	Available	Quick start
Review pull requests automatically in AL-Go for GitHub repositories	In progress in AL-Go for GitHub	AL-Go documentation once released
Apply BCQuality guidance to a development plan before coding	Preview in the plugin	Plan guidance

Quick start

Install the plugin and run your first review. You need GitHub Copilot CLI, installed and signed in, with an account and organization policy that allow its use. You do not need to clone BCQuality, build a runner, or deploy an app to Business Central.

Copilot CLI runs in any terminal, including the integrated terminal in VS Code. A plugin installed in Copilot CLI is not automatically available in the Copilot Chat panel.

1. Install the plugin:

   copilot plugin install microsoft/BCQuality
   copilot plugin list
   

   The list should include bcquality. In a Copilot CLI session, /skills list shows its two skills: al-code-review and the preview al-plan-guidance.

2. Start a new Copilot CLI session in your app folder, replacing the example path:

   cd "C:\Repos\MyBusinessCentralApp"
   copilot
   

3. Approve access only to a project you trust, then ask:

   Use the installed al-code-review skill to review the complete Business Central app in this folder without changing my source files. Return the complete BCQuality findings report.

The folder should contain app.json and your AL source; it does not need to be a Git repository. On macOS or Linux, use your app's local path instead.

What you get

Each selected review returns findings with a source location, severity, confidence, and the guidance behind them. For example, a performance finding might read:

Major · high confidence · src\CustomerExport.Codeunit.al, line 42

The country filter is evaluated inside the customer loop, so rows that will be discarded are still read from the database. Apply the filter with SetRange before calling FindSet.

Guidance: Apply filters before iterating

This example is illustrative; your host may show the same information as text, a table, or JSON. completed with no findings means nothing was flagged in that review's scope; partial or failed is not a clean result. See reading your results.

Next, try a focused request, such as reviewing performance only or reviewing your uncommitted changes. See common review requests.

PowerShell 7 (pwsh) is recommended for fast knowledge discovery. If it is unavailable, the review can still discover knowledge by reading the folders.

Documentation

I want to...	Start here
Choose direct reading, a supplied skill, or my own agent	Ways to use BCQuality
Review a file, changes, a branch, or a particular concern	Using BCQuality
Resolve setup problems, incomplete reviews, or incorrect findings	Troubleshooting and support
Browse the available guidance	Knowledge by domain
Configure the plugin or use my organization's rules	Customizing BCQuality
Contribute knowledge or improve a rule	Your first contribution
Connect a host, agent, or CI integration	Minimal integration example

All documentation and technical references.

Scope

Today's curated content covers technical AL code review and focused Supply Chain Management (SCM) and Finance functional domains. It augments the agent's judgment; it is not an exhaustive BC manual or a substitute for compilation, analyzers, tests, or human review. See coverage and limits for the available domains and the difference between a folder review and a comparison. Mechanical issues already enforced by the AL compiler or standard analyzers are intentionally left to those deterministic tools rather than duplicated here.

The SCM domain covers selected inventory, costing, reservation, tracking, and warehouse/posting workflows. The Finance domain covers selected journal posting, financial ledger, application, VAT, and posting-dimension workflows. Neither is exhaustive functional validation. Broader functional coverage such as Manufacturing, Jobs, and Service, and technologies such as PowerShell, pipelines, and Power Platform, remain valid future scope, not current coverage claims.

Preview: guidance before implementation

BCQuality can also apply its knowledge before code is written. The plugin's al-plan-guidance skill takes an existing AL development plan and returns the BCQuality constraints and validation considerations that apply to it. It is read-only: it does not create or rewrite the plan, change your code, or implement the change.

Save your plan as a file, start a new Copilot CLI session in your app folder, and ask:

Use the installed al-plan-guidance skill to enrich the plan in plans\my-change.md with applicable BCQuality guidance, without changing the plan or my source files. Return the complete guidance report.

Give the plan and the report to your coding agent together. After implementation, review the finished change with al-code-review.

This capability is a preview, and its contract may change based on pilot feedback. Integrations can also call it without the plugin; see the plan guidance example.

What's in this repo

Knowledge articles cover one concern each. Skills tell an agent how to find and apply the relevant knowledge. Both live in three layers:

Layer	Purpose
Microsoft	Microsoft-endorsed skills and their knowledge.
Community	Community-owned skills and their knowledge.
Custom	Organization-specific additions and overrides in your own fork.

All three are enabled by default; Custom is empty upstream. You do not need to configure layers to get started.

How the pieces fit

You use BCQuality through capabilities: al-code-review reviews code, and al-plan-guidance enriches a plan. Behind every capability, BCQuality's Entry router prepares the knowledge index, applies your layer and skill settings, and selects the skill that does the work. You never call Entry yourself when using the plugin.

Integrations such as CI workflows can call Entry directly with a free-text goal instead. Both paths use the same router, knowledge, and report contracts, so a plugin review and an automated review apply the same rules. Review skills return a findings-report; plan guidance returns a read-only development-guidance-report. See how agents consume BCQuality.

Versioning

Update the installed plugin from your terminal, then start a new session:

copilot plugin update bcquality

copilot plugin list shows the installed version. Release v1.8 carries plugin version 1.8.0. Installing from the repository without a tag uses the current main, which can include changes after or before the matching release. For reproducible runs and organization forks, see updates and versions.

What belongs here

Knowledge belongs here when it prevents a BC-specific mistake an otherwise capable agent would make, including false-positive findings. BC facts belong in knowledge articles, not skill instructions. See the admission test and examples.

Contributing

Partners are welcome to contribute to the layer that owns the domain, regardless of affiliation. Start with the contribution guide. To report a problem without authoring a rule, see support.

License

MIT