Skip to content
Back to search

aldc-marketplace

MarketplaceWorks with Claude CodeGitHubIncomplete

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/javiarmesto--aldc-al-development-collection/badge.svg)](/listings/javiarmesto--aldc-al-development-collection)

HTML

<a href="/listings/javiarmesto--aldc-al-development-collection"><img src="/api/listings/javiarmesto--aldc-al-development-collection/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

[Choose your surface — installation guide](https://javiarmesto.github.io/ALDC-AL-Development-Collection/start/) · [Guía en castellano](https://javiarmesto.github.io/ALDC-AL-Development-Collection/start-es/)

For the opt-in **BC29 / AL18 native profile in Copilot Chat**, see [installation, role changes and local validation](docs/native-bc29.md). BC28 remains the installer default. Claude Code and the dedicated `aldc-cli` plugin include terminal-specific BC29/AL18 contracts; see the same guide for checkout-local installation and pending runtime verification. The VS Code extension is published separately.

GitHub Copilot

Install from the VS Code Marketplace or:

code --install-extension javierarmestogonzalez.al-development-collection

Then, from the Command Palette:

- `AL Collection: Install Toolkit to Workspace` — copies the framework into your project's `.github/` - `AL Collection: Update Toolkit` — updates toolkit content; review collision behavior in your installed version - `AL Collection: Validate Installation` — verifies compliance

GitHub Copilot (Plugin Marketplace)

ALDC also ships as an installable Copilot plugin — `plugin.json` declares the agents (`agents/`) and skills (`skills/`, including the workflow skills invoked as slash commands) that get installed, for editors that support the Copilot plugin marketplace:

The root plugin retains the VS Code-oriented primitives. For **Copilot CLI**, use the dedicated terminal distribution from this checkout:

copilot plugin install ./copilot-cli-plugin
copilot plugin list

The catalog entry is `aldc-cli@aldc-marketplace`. See [plugin loading checks and limitations](docs/native-bc29.md#plugins-de-claude-code-y-copilot-cli) before testing. The directory is generated with `npm run sync:copilot-cli`.

Claude Code (Plugin)

/plugin install aldc
/aldc:al-initialize

`al-initialize` copies path-scoped rules to `.claude/rules/`, generates a project `CLAUDE.md`, and configures the workspace.

Claude Code (Direct)

Clone this repo and open it with Claude Code. The `.claude/` directory and `CLAUDE.md` are detected automatically.

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 35 lines
  1. {
  2. "name": "aldc-marketplace",
  3. "owner": {
  4. "name": "javiarmesto",
  5. "url": "https://github.com/javiarmesto"
  6. },
  7. "metadata": {
  8. "description": "ALDC — AI-Native AL Development Collection for Microsoft Dynamics 365 Business Central. Agents, skills, workflows, and coding standards for extension-only development."
  9. },
  10. "allowCrossMarketplaceDependenciesOn": ["bcquality"],
  11. "plugins": [
  12. {
  13. "name": "aldc",
  14. "source": "./claude-plugin",
  15. "description": "AL agents (incl. on-demand triage + dredd audit) + TDD subagents, skills, and workflows for BC extension development with TDD, BCQuality-cited review, event-driven architecture, and HITL gates.",
  16. "version": "5.4.0",
  17. "dependencies": [
  18. { "name": "aldc-verify" },
  19. { "name": "bcquality", "marketplace": "bcquality", "version": "^1.9.0" }
  20. ]
  21. },
  22. {
  23. "name": "aldc-extras",
  24. "source": "./claude-plugin-extras",
  25. "description": "Optional ALDC roles: technical pre-sales (estimation, SWOT, proposals) and Business Central agent building (Agent SDK). Requires aldc.",
  26. "version": "5.4.0"
  27. },
  28. {
  29. "name": "aldc-verify",
  30. "source": { "source": "github", "repo": "javiarmesto/aldc-verify-plugin", "ref": "v0.3.0" },
  31. "description": "Verifiable AL loop for Business Central: compiles AL and runs the analyzers when Claude or an ALDC implementer stops, and opens draft PRs behind a merge guard. Required by aldc.",
  32. "version": "0.3.0"
  33. }
  34. ]
  35. }
README.md · README · 756 lines
  1. <div align="center">
  2. <img src="docs/assets/images/aldc-brand-banner.svg" alt="ALDC — AL Agentic Engineering System" width="900">
  3. # ALDC — AL Agentic Engineering System
  4. **Skills-based, spec-driven, TDD-orchestrated development framework for Microsoft Dynamics 365 Business Central.**
  5. _Engineering systems, visibly reasoned._
  6. [![ALDC Core](https://img.shields.io/badge/ALDC%20Core-v1.2%20Compliant-0891B2.svg?style=flat-square&labelColor=0F172A)](docs/framework/ALDC-Core-Spec-v1.2.md)
  7. [![GitHub Release](https://img.shields.io/github/v/release/javiarmesto/ALDC-AL-Development-Collection?style=flat-square&labelColor=0F172A&color=D946EF&label=GitHub%20release)](https://github.com/javiarmesto/ALDC-AL-Development-Collection/releases)
  8. [![VS Code Marketplace](https://img.shields.io/visual-studio-marketplace/v/javierarmestogonzalez.al-development-collection?style=flat-square&labelColor=0F172A&color=0891B2&label=Marketplace)](https://marketplace.visualstudio.com/items?itemName=javierarmestogonzalez.al-development-collection)
  9. [![Claude Code Plugin](https://img.shields.io/badge/Claude%20Code-plugin%20available-38BDF8.svg?style=flat-square&labelColor=0F172A)](claude-plugin/)
  10. [![Framework](https://img.shields.io/badge/framework-AI--Native--Instructions-0891B2?style=flat-square&labelColor=0F172A)](https://danielmeppiel.github.io/awesome-ai-native/)
  11. [![License](https://img.shields.io/badge/license-MIT-0891B2?style=flat-square&labelColor=0F172A)](./LICENSE)
  12. [![GitHub Issues](https://img.shields.io/github/issues/javiarmesto/ALDC-AL-Development-Collection?style=flat-square&labelColor=0F172A&color=0891B2)](https://github.com/javiarmesto/ALDC-AL-Development-Collection/issues)
  13. [![GitHub Stars](https://img.shields.io/github/stars/javiarmesto/ALDC-AL-Development-Collection?style=flat-square&labelColor=0F172A&color=D946EF)](https://github.com/javiarmesto/ALDC-AL-Development-Collection/stargazers)
  14. </div>
  15. ALDC combines specialist agents, reusable domain skills and human approval to
  16. support Business Central extension development from requirements to review.
  17. **ALDC 5.3.0** ships the ten workflows as skills instead of prompt files: you run
  18. `/al-spec-create`, `/al-build` and the rest as slash commands, because VS Code's Copilot
  19. harness does not load `.prompt.md` files. It builds on 5.2.1, which lets Architect, Spec
  20. and Conductor launch their internal subagents in that harness and the Copilot CLI, and on
  21. 5.2.0: explicit security review, BCQuality article dispositions, stronger planning and
  22. delegation contracts, and configurable Conductor autonomy and test execution.
  23. The Marketplace badge reports the currently published extension version.
  24. See [what changed in 5.3.0](#what-changed-in-530), [in 5.2.1](#what-changed-in-521), [in 5.2.0](#what-changed-in-520) and the [changelog](CHANGELOG.md).
  25. | Channel | Current delivery | Scope |
  26. | --- | --- | --- |
  27. | [VS Code Marketplace](https://marketplace.visualstudio.com/items?itemName=javierarmestogonzalez.al-development-collection) | 5.3.0 (the badge reports the live version) | Copilot Chat extension and packaged toolkit; existing projects adopt new contracts through Update Toolkit. |
  28. | [Latest GitHub release](https://github.com/javiarmesto/ALDC-AL-Development-Collection/releases/tag/v5.3.0) | v5.3.0 | Tagged source, toolkit archive and VSIX file; a separate delivery from the Marketplace listing. |
  29. | Source at tag `v5.3.0` | Root package, Copilot catalogs and generated CLI/Codex distributions 5.3.0 | Used from a checkout by the terminal hosts. The npm registry package is not maintained: from a checkout or a local `.tgz`, run `npx --no aldc …`. |
  30. | [Claude Code edition](claude-plugin/) | 5.3.0 | Maintained independently, with its own agents, hooks, contracts and release evolution. |
  31. Claude Code remains part of ALDC, with a shared engineering foundation and its own
  32. maintained agents, hooks, contracts and release evolution. Its surface changes are
  33. documented separately in the [changelog](CHANGELOG.md) and
  34. [`claude-plugin/SURFACE.md`](claude-plugin/SURFACE.md). A version in one channel
  35. does not establish an equivalent update in another, or publication to npm.
  36. For an installed Claude Code plugin, refresh the official source and plugin:
  37. ```text
  38. claude plugin marketplace update aldc-marketplace
  39. claude plugin update aldc@aldc-marketplace
  40. ```
  41. Open a new Claude Code session and check the version against its plugin catalog. This refreshes the
  42. plugin's contracts; project-local toolkit copies need their own reviewed update.
  43. See [installation and recovery](docs/plugin-packaging.md) for each surface.
  44. ## Start here
  45. Choose the [installation guide](https://javiarmesto.github.io/ALDC-AL-Development-Collection/start/) for your host, then follow the [Quick Start](#quick-start). Use your own AL project and BC sandbox; check its `app.json`, symbols and permissions before any build or publication. ALDC supplies development context and workflows, not a Business Central environment.
  46. Start with the toolkit installed in your project and a bounded requirement. Architect and Spec prepare the design and technical specification for your review before implementation.
  47. [Architecture and distribution map](ARCHITECTURE.md) · [Installation/recovery](docs/plugin-packaging.md) · [Doctor](tools/context-doctor/README.md) · [Public BC/AL/MCP repository catalogue](docs/public-repositories.md).
  48. ## What changed in 5.3.0
  49. The ten workflows are now skills at `skills/<name>/SKILL.md`, run as slash commands:
  50. `/al-spec-create`, `/al-build`, `/al-pr-prepare`, `/al-memory-create`, `/al-context-create`,
  51. `/al-initialize`, `/al-agent-create`, `/al-agent-task`, `/al-agent-instructions` and
  52. `/al-agent-test`. Names that had a dot now have a hyphen (`al-spec.create` is
  53. `/al-spec-create`), and `@workspace use <name>` is no longer how you call them.
  54. `al-agent.build-instructions` is gone; it was a copy of `al-agent.instructions`.
  55. A workflow runs with the tools of the agent that invokes it. In existing projects, Update
  56. Toolkit removes the `.github/prompts/` files it installed and keeps a backup; a prompt
  57. you edited stays in place and is reported on every update until you replace it.
  58. ## What changed in 5.2.1
  59. The Copilot engine used by VS Code's Copilot harness and the Copilot CLI cannot launch
  60. an agent marked `disable-model-invocation`, so the four internal subagents no longer
  61. use that field. They stay out of the agent picker, and each one now answers
  62. `NOT_FOR_THIS_CALLER` unless its brief starts with the `Caller:` line that Architect,
  63. Spec or Conductor writes. Copilot CLI agents use `claude-sonnet-5.5`, and its plugin
  64. agents are addressed as `aldc-cli:<name>`. In existing projects, run Update Toolkit to
  65. adopt these contracts.
  66. ## What changed in 5.2.0
  67. Reviews now include five mandatory security checks and a disposition for every
  68. selected BCQuality article: apply it or explain why it does not apply. Agent SDK
  69. reviews check completeness when the app extends `Agent Metadata Provider`.
  70. Architect, Spec Agent and Planning Subagent cite the knowledge informing their
  71. decisions. Planning records the standard publisher flow and the chosen event;
  72. subscribers explain the choice above their attribute.
  73. Conductor delegates research, each implementation phase and review to their
  74. owners. It records a Mini-design for eligible LOW work without an architecture
  75. document; the Core v1.2 architecture gate for MEDIUM/HIGH remains. Final reports
  76. distinguish verification performed, checks pending, decisions and open risks.
  77. `aldc.yaml` adds `autonomy.gates: human | pre-approved` and `testing.runner`.
  78. Human is the default. With the requester's prior approval, gates become recorded
  79. checkpoints in `decisions.md`; hard stops still produce `blocked.md`.
  80. `autonomy.commit` remains `false` by default. A configured test runner supplies
  81. actual per-test results after compilation; without one, execution is `not-run`.
  82. `ALDC_GATES` and `ALDC_TEST_RUNNER` can override the corresponding settings where
  83. the host exposes environment variables. These contracts add no host hooks.
  84. The companion VS Code Project Manager offers an explicit
  85. BCQuality mode selector for existing projects and native VS Code installation
  86. help. Fresh extension installations select BCQuality plugin mode.
  87. The canonical `aldc.yaml` keeps its `external-multiroot` default. Project Manager's
  88. Human/default and Automatic options map to `autonomy.gates: human` and
  89. `pre-approved`; Automatic requires prior approval and grants no additional commit
  90. or publication permission.
  91. Fresh installations in both Chat profiles select BCQuality plugin mode with
  92. automatic discovery and no fixed plugin version pin. Updates preserve the
  93. project's selected mode, availability preference and provider identity; customized
  94. configuration follows the existing collision policy. A newly seeded workspace
  95. includes an external knowledge folder only for an enabled external mode and uses
  96. its configured path. Existing workspace files remain owned by the project.
  97. Both Chat profiles retain 5.0.4's dependency guidance: resolve only libraries used
  98. by the planned tests from approved dependencies, declared versions and symbols.
  99. At 5.2.0, Claude Code stayed on its independently maintained 5.1.1 edition; it is 5.3.0 today.
  100. ### Use Project Manager in VS Code
  101. 1. Update the extension to **5.3.0** and run **AL Collection: Open Project Manager**.
  102. 2. Select your solution, preview its toolkit installation or update, review customizations and confirm. Updating the extension does not replace the toolkit already installed in a project.
  103. 3. In **BCQuality**, choose **Plugin — recommended**, **External knowledge folder** or **Disabled — native review**, then select **Change mode**. **Install BCQuality…** copies the official repository URL for the native source prompt; paste it there. Hosts without that command open the installation guide. **Manage plugins** opens the host's inventory.
  104. 4. In **Workflow autonomy**, choose **Human — approve each checkpoint** or **Automatic — prior approval**, then select **Apply mode**. Start a new chat with **AL Development Conductor**.
  105. Human waits at approval checkpoints. Automatic records those decisions and continues within the approved task, retaining hard stops. Changes preserve commit and publication permissions and keep a recoverable configuration copy. A retained older Conductor must be updated before using the autonomy control. Plugin installation and loading are verified by the host.
  106. These Project Manager controls apply to VS Code/Copilot Chat. The other hosts keep their own configuration and permissions. Canvas remains a separate experimental companion.
  107. ### Earlier Copilot Chat improvements
  108. **5.0.3 — knowledge before implementation.** Architect and the independent Spec
  109. Agent consult BCQuality through `al-knowledge`, read complete articles and record
  110. the evidence used in their decisions. Hosts without a skill invocation tool follow
  111. the instructions inline. BCQuality is installed separately; unavailable knowledge
  112. and manual fallbacks remain explicit. Windows validation and generated-path
  113. handling also improve.
  114. **5.0.4 — native AL tools for your project.** Copilot Chat's `bc29-native`
  115. Developer and Implementation Subagent can query object IDs within approved project
  116. ranges. IDs are not reserved, and the local scan excludes installed extensions.
  117. Architect/Spec research stays bounded to relevant symbols and dependencies;
  118. native diagnostics use the compatible AL18.1 selector. Developer can compile the
  119. selected AL project and consult its diagnostics as part of the implementation
  120. workflow.
  121. Conductor keeps coordination and Reviewer keeps query permissions. Publication
  122. requires explicit authorization. Graph, LSP, profiling and snapshot
  123. operations remain optional and depend on the installed host and tools. The
  124. [native tooling guide](docs/framework/native-al-tools.md) explains the available
  125. capabilities, project selection and permissions. Claude Code's later surface
  126. releases follow their own contracts.
  127. To update an existing VS Code project, update the extension, run **AL Collection:
  128. Update Toolkit**, review the preview and managed-file collisions, then run
  129. **AL Collection: Verify Installed Toolkit** and reload the window. Each terminal
  130. plugin has its own update route. Earlier improvements, including Copilot Chat's
  131. Sonnet 5.5 selection, remain documented under their original changelog versions.
  132. | Capability | What it provides |
  133. | --- | --- |
  134. | Architect and Spec Agent | Approved design, bounded Specs, explicit dependencies and joint consistency review before implementation. |
  135. | Conductor | Planning, implementation and review coordinated through existing human gates. |
  136. | Doctor | Read-only diagnosis of configuration and operation-specific runtime observations. |
  137. | Recommended BCQuality | Plugin or external multiroot reviews, with discovery, loading, execution and index evidence distinguished. |
  138. | BC28 / BC29-native | Explicit Chat profile selection with role-specific tools and AL18 guidance. |
  139. | Host adapters | Dedicated content for Copilot Chat, Copilot CLI, Claude Code and Codex. |
  140. | Recoverable installation | Preview, visible collisions, receipts, verification and restoration. |
  141. See [Spec Agent](docs/spec-agent.md), [profiles](docs/native-bc29.md),
  142. [plugins and recovery](docs/plugin-packaging.md) and [Doctor](tools/context-doctor/README.md).
  143. ---
  144. ## Why ALDC?
  145. AI accelerates AL development — but raw code generation is unpredictable, hard to review, and easy to get wrong against Business Central's extension rules. ALDC adds the engineering discipline around it.
  146. | Without ALDC | With ALDC |
  147. | --- | --- |
  148. | Ad-hoc, one-shot code generation | **Contract-driven** development (architecture → spec → human approval → test-plan → code) |
  149. | No checkpoints — you find out at the end | **Human-in-the-loop gates** at every phase |
  150. | Tests written last (or never) | **TDD-orchestrated** — tests FIRST, then code |
  151. | Base-app edits sneak in | **Extension-only** by construction |
  152. | "Looks fine to me" reviews | **Cited reviews & audits** against BC knowledge |
  153. ---
  154. ## What is ALDC?
  155. ALDC (AL Development Collection) transforms how you develop Business Central extensions. Instead of ad-hoc code generation, it provides **structured, contract-driven development** with specialized agents, composable skills, and human-in-the-loop gates — with host-specific distributions:
  156. - **GitHub Copilot** — Agents, skills (workflows included) and instructions in `.github/` and `agents/`
  157. - **Claude Code** — Agents, skills, rules, and hooks in `.claude/` and `claude-plugin/`
  158. - **Copilot CLI** — Terminal agents and commands in `copilot-cli-plugin/`
  159. - **Codex** — Generated agent profiles and a local ALDC skill in `plugins/aldc-codex/`
  160. ---
  161. ## Updating ALDC
  162. The VS Code extension and each host plugin are separate distribution channels.
  163. Updating the VSIX does not refresh installed Claude Code, Copilot CLI or Codex plugins.
  164. In VS Code, open **AL Collection: Open Project Manager** (also available as a view in
  165. the Explorer) to preview and confirm the update, verify the installation, restore the
  166. previous transaction and run Doctor without leaving the panel; the **ALDC Visor** view
  167. lists the plan artifacts of each requirement with template-specific icons. The palette
  168. command **AL Collection: Update Toolkit** remains available: review file collisions,
  169. verify the installation and reload the window. Existing installations retain their
  170. recorded profile. The extension's optional `al-collection.autoInstall` setting is
  171. `false` by default; when enabled, it attempts installation on AL project detection.
  172. Toolkit restoration covers installation files, not Business Central data or all
  173. project history. See [installation and recovery](docs/plugin-packaging.md).
  174. ## Installation
  175. [Choose your surface — installation guide](https://javiarmesto.github.io/ALDC-AL-Development-Collection/start/) · [Guía en castellano](https://javiarmesto.github.io/ALDC-AL-Development-Collection/start-es/)
  176. For the opt-in **BC29 / AL18 native profile in Copilot Chat**, see
  177. [installation, role changes and local validation](docs/native-bc29.md).
  178. BC28 remains the installer default. Claude Code and the dedicated `aldc-cli` plugin
  179. include terminal-specific BC29/AL18 contracts; see the same guide for checkout-local
  180. installation and pending runtime verification. The VS Code extension is published separately.
  181. ### GitHub Copilot
  182. Install from the VS Code Marketplace or:
  183. ```bash
  184. code --install-extension javierarmestogonzalez.al-development-collection
  185. ```
  186. Then, from the Command Palette:
  187. - `AL Collection: Install Toolkit to Workspace` — copies the framework into your project's `.github/`
  188. - `AL Collection: Update Toolkit` — updates toolkit content; review collision behavior in your installed version
  189. - `AL Collection: Validate Installation` — verifies compliance
  190. ### GitHub Copilot (Plugin Marketplace)
  191. ALDC also ships as an installable Copilot plugin — `plugin.json` declares the agents (`agents/`) and skills (`skills/`, including the workflow skills invoked as slash commands) that get installed, for editors that support the Copilot plugin marketplace:
  192. The root plugin retains the VS Code-oriented primitives. For **Copilot CLI**, use
  193. the dedicated terminal distribution from this checkout:
  194. ```bash
  195. copilot plugin install ./copilot-cli-plugin
  196. copilot plugin list
  197. ```
  198. The catalog entry is `aldc-cli@aldc-marketplace`.
  199. See [plugin loading checks and limitations](docs/native-bc29.md#plugins-de-claude-code-y-copilot-cli)
  200. before testing. The directory is generated with `npm run sync:copilot-cli`.
  201. ### Claude Code (Plugin)
  202. ```bash
  203. /plugin install aldc
  204. /aldc:al-initialize
  205. ```
  206. `al-initialize` copies path-scoped rules to `.claude/rules/`, generates a project `CLAUDE.md`, and configures the workspace.
  207. ### Claude Code (Direct)
  208. Clone this repo and open it with Claude Code. The `.claude/` directory and `CLAUDE.md` are detected automatically.
  209. ---
  210. ## Quick Start
  211. ### GitHub Copilot
  212. 1. Install the extension and open your AL project
  213. 2. Run `AL Collection: Install Toolkit to Workspace`
  214. 3. Start with `/al-spec-create` plus your requirement
  215. 4. Follow the guided flow
  216. ### Claude Code
  217. 1. `/plugin install aldc`
  218. 2. `/aldc:al-initialize`
  219. 3. Call any agent: `@al-architect`, `@al-developer`, `@al-conductor`, `@al-presales`
  220. 4. Or run a workflow: `/aldc:al-spec-create`
  221. See [QUICKSTART.md](docs/framework/QUICKSTART.md) for the full onboarding guide.
  222. ---
  223. ## Key Features
  224. ### 🤖 5 Core Agents — one specialist per development phase
  225. | Agent | Role |
  226. | --- | --- |
  227. | `@AL Architecture & Design Specialist` | Designs solutions, information flows, technical decisions |
  228. | `@AL Spec Agent` | Specifies approved designs through the shared `al-spec-create` contract |
  229. | `@AL Implementation Specialist` | Implements, debugs, quick adjustments |
  230. | `@AL Development Conductor` | Orchestrates TDD implementation with subagents |
  231. | `@AL Pre-Sales & Project Estimation Specialist` | Estimation and scoping |
  232. ### 🔬 4 Internal Subagents — launched by Architect, Spec Agent or Conductor, never picked directly
  233. - **AL Planning Subagent** — research and context gathering
  234. - **AL Implementation Subagent** — TDD-only (tests FIRST, code SECOND)
  235. - **AL Code Review Subagent** — review against spec + architecture
  236. - **AL Knowledge Consultant** — one BCQuality knowledge consultation per question, for Architect, Spec Agent and Conductor
  237. ### 🩺 3 On-demand Specialists — user-invocable, read-only on code
  238. - `@AL Triage` — reactive diagnosis: reproduce → root-cause → minimal-fix recommendation
  239. - `@Dredd` — independent auditor: BCQuality-cited static audit with an advisory verdict
  240. - `@AL Developer Reviewer` — independent review of direct Developer increments
  241. ### 🧠 Core domain skills — knowledge loaded on demand
  242. - **Required:** api · copilot · debug · performance · events · permissions · testing
  243. - **Recommended:** migrate · pages · translate · estimation
  244. - The complete source has 16 skill directories, including contribution assistance, BC Agent modules and the shared AL review pipeline. See [skills/](skills/). Host distributions also package role entries and workflows; their folder counts measure a different scope.
  245. ### ⚙️ 10 Workflows — skills run as slash commands
  246. - **Core (6):** `/al-spec-create` · `/al-build` · `/al-pr-prepare` · `/al-context-create` · `/al-memory-create` · `/al-initialize`
  247. - **BC Agent Builder (4, optional):** `/al-agent-create` · `/al-agent-task` · `/al-agent-instructions` · `/al-agent-test`
  248. ### 📐 Scoped instructions and the Copilot entrypoint
  249. Nine `*.instructions.md` files apply by their declared globs, including the security rule and the optional BC Agent toolkit rule. `copilot-instructions.md` is the always-on entrypoint and `index.md` is documentation; neither is an additional scoped rule. See [instructions/](instructions/).
  250. ### 📚 BCQuality (recommended) — external, citable BC knowledge layer
  251. Optional cited review through an explicitly selected host plugin or external multiroot knowledge base. Agents distinguish discovery, loading, execution and best-effort index generation; unavailable providers retain native A–G coverage. See [`docs/bcquality.md`](docs/bcquality.md).
  252. ### 📄 Contracts per Requirement — structured docs in `.github/plans/{req_name}/`
  253. - `{req_name}.architecture.md` — solution design (from architect)
  254. - `{req_name}.spec.md` — technical blueprint (from `/al-spec-create`)
  255. - `{req_name}.test-plan.md` — test strategy
  256. - `memory.md` — global context across sessions
  257. ---
  258. ## How It Works
  259. ### Development Flow
  260. ```text
  261. LOW complexity:
  262. al-spec-create → @AL Implementation Specialist
  263. MEDIUM/HIGH complexity:
  264. @AL Architecture & Design Specialist → al-spec-create → @AL Development Conductor
  265. ```
  266. The architect designs the solution and can decompose complex requirements into multiple specs, each implemented independently by the conductor.
  267. ```mermaid
  268. flowchart TD
  269. REQ[Requirement] --> CLASSIFY{Complexity?}
  270. CLASSIFY -->|LOW| SPEC_LOW[al-spec-create]
  271. SPEC_LOW --> DEV["@AL Implementation Specialist"]
  272. CLASSIFY -->|MEDIUM/HIGH| ARCH["@AL Architecture & Design Specialist"]
  273. ARCH -->|Designs solution| ARCH_DOC[architecture.md]
  274. ARCH --> DECOMPOSE{Decompose?}
  275. DECOMPOSE -->|Yes| SPEC_A[al-spec-create → spec-A]
  276. DECOMPOSE -->|Yes| SPEC_B[al-spec-create → spec-B]
  277. DECOMPOSE -->|No| SPEC_SINGLE[al-spec-create → spec.md]
  278. SPEC_A --> COND_A["@AL Development Conductor"]
  279. SPEC_B --> COND_B["@AL Development Conductor"]
  280. SPEC_SINGLE --> COND["@AL Development Conductor"]
  281. ```
  282. ### TDD Orchestration
  283. The conductor enforces Test-Driven Development:
  284. ```mermaid
  285. flowchart LR
  286. PLAN[Phase 1\nPlanning] --> IMPL[Phase N\nImplementation]
  287. IMPL --> REVIEW[Review\nSubagent]
  288. REVIEW --> HITL{HITL Gate}
  289. HITL -->|Approved| NEXT[Next Phase]
  290. HITL -->|Revise| IMPL
  291. subgraph IMPL_DETAIL[Implement Subagent - TDD]
  292. RED[RED\nWrite failing tests] --> GREEN[GREEN\nMinimal code]
  293. GREEN --> REFACTOR[REFACTOR\nAL patterns]
  294. end
  295. ```
  296. 1. Planning subagent researches context
  297. 2. Implement subagent creates tests FIRST (RED)
  298. 3. Implement subagent writes code to pass tests (GREEN)
  299. 4. Implement subagent refactors to AL patterns (REFACTOR)
  300. 5. Review subagent validates against spec + architecture
  301. 6. Human approves each phase (HITL gate)
  302. ### Framework Architecture
  303. ```mermaid
  304. graph TB
  305. subgraph PUBLIC["Public Agents (user-invocable)"]
  306. ARCH["@AL Architecture & Design Specialist"]
  307. SPEC["@AL Spec Agent"]
  308. DEV["@AL Implementation Specialist"]
  309. COND["@AL Development Conductor"]
  310. PRE["@AL Pre-Sales & Project Estimation Specialist"]
  311. end
  312. subgraph INTERNAL["Internal Subagents (not user-invocable)"]
  313. PLAN[AL Planning Subagent]
  314. IMPL[AL Implementation Subagent]
  315. REV[AL Code Review Subagent]
  316. KNOW[AL Knowledge Consultant]
  317. end
  318. subgraph SKILLS["Knowledge Skills (11 of 16 shown)"]
  319. S1[skill-api]
  320. S2[skill-copilot]
  321. S3[skill-debug]
  322. S4[skill-performance]
  323. S5[skill-events]
  324. S6[skill-permissions]
  325. S7[skill-testing]
  326. S8[skill-migrate]
  327. S9[skill-pages]
  328. S10[skill-translate]
  329. S11[skill-estimation]
  330. end
  331. subgraph WORKFLOWS["Core Workflows (6 of 10)"]
  332. W1[al-spec-create]
  333. W2[al-build]
  334. W3[al-pr-prepare]
  335. W4[al-context-create]
  336. W5[al-memory-create]
  337. W6[al-initialize]
  338. end
  339. ARCH --> SKILLS
  340. DEV --> SKILLS
  341. COND --> INTERNAL
  342. PRE --> SKILLS
  343. W1 --> SPEC
  344. W1 --> ARCH
  345. W1 --> COND
  346. ```
  347. ### Contract Structure
  348. ```text
  349. .github/
  350. └── plans/
  351. ├── memory.md ← Global (cross-session context)
  352. └── {req_name}/
  353. ├── {req_name}.architecture.md ← From @AL Architecture & Design Specialist
  354. ├── {req_name}.spec.md ← From al-spec-create
  355. ├── {req_name}.test-plan.md ← From al-spec-create or conductor
  356. ├── {req_name}-plan.md ← From @AL Development Conductor (Planning)
  357. ├── {req_name}-phase-1-complete.md
  358. └── {req_name}-phase-N-complete.md
  359. ```
  360. ---
  361. ## Routing Guide
  362. | Complexity | Route | When |
  363. | ---------- | ----- | ---- |
  364. | **LOW** | `al-spec-create` → `@AL Implementation Specialist` | Simple field, validation, single UI change |
  365. | **MEDIUM** | `@AL Architecture & Design Specialist` → `al-spec-create` → `@AL Development Conductor` | Business logic, event-driven feature |
  366. | **HIGH** | `@AL Architecture & Design Specialist` → `al-spec-create` → `@AL Development Conductor` | Multi-module, external integration, architectural change |
  367. **Not sure where to start?**
  368. ```text
  369. @AL Architecture & Design Specialist
  370. I need to [describe your requirement]
  371. ```
  372. The architect analyzes requirements, designs the solution, and recommends the appropriate workflow.
  373. ---
  374. ## ALDC for Claude Code
  375. ALDC is available as a native **Claude Code** integration in two forms:
  376. - **Official Plugin** (`claude-plugin/`) — Install with `/plugin install aldc`, namespaced as `aldc:*`
  377. - **Direct Integration** (`.claude/`) — Auto-detected when opening the repo in Claude Code
  378. ### What's Included
  379. | Primitive | Direct (`.claude/`) | Plugin (`aldc:`) | Count |
  380. | --------- | ------------------- | ---------------- | ----- |
  381. | Agents | `.claude/agents/` | `agents/` | 9 public + 4 internal |
  382. | Skills | `.claude/skills/` | `skills/` | 16 knowledge modules + 9 role entries + 10 workflows |
  383. | Rules | `.claude/rules/` | `rules/` (injected via `al-initialize`) | 9 coding standards |
  384. | MCP Servers | `.mcp.json` | `.mcp.json` | 3 servers |
  385. | Hooks | `.claude/settings.json` | `hooks/hooks.json` | 2 reminders (Direct); SessionStart, PreToolUse, PostToolUse and Stop hooks (Plugin) |
  386. | Instructions | `CLAUDE.md` | `CLAUDE.md` | Agent routing, workflows |
  387. ### How It Maps
  388. ```text
  389. GitHub Copilot → Claude Code (Direct) → Claude Code (Plugin)
  390. ──────────────────────────────────────────────────────────────────────────────────────
  391. agents/*.agent.md → .claude/agents/*.md → agents/*.md
  392. skills/*/SKILL.md → .claude/skills/*/SKILL.md → skills/*/SKILL.md
  393. instructions/*.md → .claude/rules/*.md → rules/*.md
  394. skills/al-*/SKILL.md → .claude/skills/ (workflows) → skills/ (workflows)
  395. .github/copilot-instructions.md → CLAUDE.md → plugin.json + CLAUDE.md
  396. ```
  397. ### Agent Routing (Claude Code)
  398. | Agent | Direct | Plugin |
  399. | ----- | ------ | ------ |
  400. | Architecture & Design | `@al-architect` | `@aldc:al-architect` |
  401. | Technical Specification | `@al-spec-agent` | `@aldc:al-spec-agent` |
  402. | Implementation | `@al-developer` | `@aldc:al-developer` |
  403. | TDD Orchestration | `@al-conductor` | `@aldc:al-conductor` |
  404. | Estimation & Scoping | `@al-presales` | `@aldc:al-presales` |
  405. | Agent Builder | `@al-agent-builder` | `@aldc:al-agent-builder` |
  406. ### Workflows (Claude Code)
  407. | Workflow | Direct | Plugin |
  408. | -------- | ------ | ------ |
  409. | Create specifications | `/al-spec-create` | `/aldc:al-spec-create` |
  410. | Build & deploy | `/al-build` | `/aldc:al-build` |
  411. | Prepare PR | `/al-pr-prepare` | `/aldc:al-pr-prepare` |
  412. | Session memory | `/al-memory-create` | `/aldc:al-memory-create` |
  413. | Project context | `/al-context-create` | `/aldc:al-context-create` |
  414. | Environment setup | `/al-initialize` | `/aldc:al-initialize` |
  415. ### Hooks
  416. The Direct integration (`.claude/settings.json`) has two reminder hooks:
  417. - **PostToolUse** (Write/Edit) — Reminds to run tests after file modifications
  418. - **Stop** — Reminds to verify Skills Evidencing was declared
  419. The plugin ships its own hook set in [`claude-plugin/hooks/hooks.json`](claude-plugin/hooks/hooks.json): SessionStart, PreToolUse, PostToolUse and Stop.
  420. ### Plugin User Configuration
  421. On first enable, the plugin prompts for optional settings:
  422. | Setting | Description |
  423. | ------- | ----------- |
  424. | `bcSandboxUrl` | URL of your Business Central sandbox environment |
  425. | `publisherName` | Your extension publisher name for app.json |
  426. ---
  427. ## Using BCQuality (recommended)
  428. BCQuality supports explicit `plugin` and `external-multiroot` modes in `aldc.yaml`.
  429. Plugin mode loads the configured skill (default `al-code-review`); multiroot
  430. retains the external Entry workflow. An expected version or commit is not proof of
  431. the installed identity, and catalog discovery is not execution. When unavailable,
  432. native review continues. See [configuration and evidence](docs/bcquality.md).
  433. ## BC Agent Builder (optional)
  434. Build Business Central Agents with the AI Development Toolkit and Agent SDK.
  435. Includes: `@AL Agent Builder` agent, 3 skills, 4 workflows, validation tools.
  436. See [BC Agent Builder documentation](docs/bc-agent-builder.md).
  437. ---
  438. ## Validation
  439. ```bash
  440. node tools/aldc-validate/index.js --config aldc.yaml
  441. ```
  442. Expected result: a final line starting with `✅ ALDC Core v1.1 COMPLIANT`. The validator's banner still reads v1.1; it checks the Core v1.2 manifest in `aldc.yaml`.
  443. ---
  444. ## File Structure
  445. ```text
  446. ALDC-AL-Development-Collection/
  447. │
  448. │── GitHub Copilot ─────────────────────────────────────
  449. ├── .github/
  450. │ ├── copilot-instructions.md # Master coordination
  451. │ └── plans/ # Per-requirement contracts
  452. │ ├── memory.md # Global memory (cross-session)
  453. │ └── {req_name}/
  454. │ ├── {req_name}.architecture.md
  455. │ ├── {req_name}.spec.md
  456. │ └── {req_name}.test-plan.md
  457. ├── agents/ # 13 agents (5 core + 3 on-demand + 4 subagents + 1 extension)
  458. ├── skills/ # 16 knowledge skills + 10 workflow skills (slash commands)
  459. ├── instructions/ # 9 scoped rules + entrypoint (10 contracts); documentation excluded
  460. │
  461. │── Claude Code (Direct) ───────────────────────────────
  462. ├── CLAUDE.md # Master instructions
  463. ├── .mcp.json # MCP server configuration
  464. ├── .claude/
  465. │ ├── agents/ # 13 agents (9 public + 4 internal)
  466. │ ├── skills/ # 9 role entries + 10 workflows + 16 knowledge
  467. │ ├── rules/ # 9 path-scoped coding standards
  468. │ └── settings.json # Hooks + permissions
  469. │
  470. │── Claude Code Plugin ─────────────────────────────────
  471. ├── claude-plugin/ # independently maintained Claude Code edition
  472. │ ├── .claude-plugin/plugin.json # Plugin manifest
  473. │ ├── agents/ # 13 agents (auto-discovered)
  474. │ ├── skills/ # 9 role entries + 10 workflows + 16 knowledge
  475. │ ├── hooks/hooks.json # SessionStart, PreToolUse, PostToolUse and Stop hooks
  476. │ ├── rules/ # 9 rules (injected via al-initialize)
  477. │ ├── .mcp.json # 3 MCP servers
  478. │ └── README.md # Plugin documentation
  479. │
  480. │── Shared ─────────────────────────────────────────────
  481. ├── docs/
  482. │ ├── framework/ # Normative spec + diagrams
  483. │ └── templates/ # Immutable contract templates (7)
  484. ├── tools/aldc-validate/ # ALDC Core validator
  485. ├── aldc.yaml # Core v1.2 configuration
  486. ├── CHANGELOG.md # Version history
  487. └── README.md # This file
  488. ```
  489. ---
  490. ## Reference
  491. ### Framework Documentation
  492. - [Core Specification v1.2](docs/framework/ALDC-Core-Spec-v1.2.md)
  493. - [Architecture Diagrams](docs/framework/ALDC-Architecture-Diagrams.md)
  494. - [Manifesto](docs/framework/ALDC-Manifesto.md)
  495. - [Quickstart](docs/framework/QUICKSTART.md)
  496. - [Governance](docs/framework/ALDC-Governance.md)
  497. - [Compliance Model](docs/framework/ALDC-Compliance-Model.md)
  498. - [Migration Guide v1.0→v1.1](docs/framework/ALDC-Migration-v1.0-to-v1.1.md)
  499. ### MCP Servers Integration
  500. | Server | Purpose |
  501. | ------ | ------- |
  502. | [al-symbols-mcp](https://github.com/StefanMaron/AL-Dependency-MCP-Server) | AL object analysis from compiled .app packages |
  503. | [context7](https://github.com/upstash/context7) | Up-to-date library documentation retrieval |
  504. | [microsoft-docs](https://learn.microsoft.com/training/support/mcp) | Official Microsoft/Azure documentation search |
  505. ### Requirements
  506. **GitHub Copilot**
  507. - Visual Studio Code 1.85.0+
  508. - GitHub Copilot (agent and skill features)
  509. - AL Language Extension
  510. - Node.js 20+ (for validator)
  511. **Claude Code**
  512. - Claude Code CLI v1.0.33+
  513. - AL Language Extension
  514. - Node.js 20+ (for MCP servers via npx)
  515. ---
  516. ## What's New
  517. ### 5.0.0 — one delivery across every surface
  518. Dedicated Spec Agent, per-operation Doctor, BC29-native tooling, terminal adapters
  519. and recoverable initialization, plus the VS Code Project Manager and Visor, BCQuality
  520. read as context in the design phases, and a Claude Code plugin that is generated
  521. rather than hand-written. It supersedes the unreleased 4.3.0 and 4.3.1, neither of
  522. which was ever tagged or published. Major for the plugin layout and the per-surface
  523. plans and audits roots; see the [5.0.0 changelog](CHANGELOG.md).
  524. ### 4.2.0 — Conformance release
  525. The framework now enforces its own spec in CI.
  526. - **Core Spec v1.2 (original 4.2.0 release)** — originally normalized the tier model to 4 core agents + 2 on-demand (`al-triage`, `dredd`) + 3 subagents + 1 extension (`al-agent-builder`); 16 skills; 11 workflows. The subsequent canonical Spec Agent increment adds a fifth core role (11 total); Developer Reviewer adds a direct review role (12 total, with a shared review pipeline).
  527. - **Conformance tooling** — `scripts/check-conformance.js` (counters, cross-references, links, frontmatter) and `scripts/sync-foundation.js --check` (zero drift between the canonical trees and `packages/foundation/`) run on every push and PR.
  528. - **`ARCHITECTURE.md`** — one-page map of what is source, what is generated, and which distribution channel consumes each tree.
  529. - Fixed: truncated `skill-manifest` in `packages/foundation/`, broken README links, undeclared primitives in `aldc.yaml`, contradictory counters.
  530. ### 4.1.0 — Lower token cost & cited audits
  531. - **⚡ Lower token / AIC cost** — trimmed always-on entrypoint (~31% lighter), narrow instruction globs (`applyTo` by object type), curated context passing, condensed primitives, BCQuality task-context built once and passed inline.
  532. - **📚 Cited reviews & audits with BCQuality (recommended)** — agents back findings with a pinned BC knowledge base; graceful native fallback (never blocks).
  533. - **`@AL Triage`** and **`@Dredd`** — read-only on-demand specialists.
  534. - **`skill-contribution-assistant`** — guided contribution workflow.
  535. - Restored full architecture & spec templates with authoring guidance.
  536. ### 4.0.0 — Token efficiency & foundation layout
  537. - Agents, instructions, skills, prompts, and templates condensed for a lower token footprint — behavior preserved.
  538. - New `packages/foundation/` layout; Architecture Decision Records (`docs/decisions/`); new `al-agent-instructions` prompt; English-only content.
  539. - **Breaking:** primitives also available under `packages/foundation/`; agent/instruction/skill wording token-optimized (behavior unchanged).
  540. See [CHANGELOG.md](CHANGELOG.md) for full details.
  541. ---
  542. ## Author
  543. **Javier Armesto González**
  544. Microsoft MVP (Business Central & Azure AI Services)
  545. Head of R&D & AI at VS Sistemas
  546. [LinkedIn](https://www.linkedin.com/in/jarmesto/) · [Tech Sphere Dynamics](https://techspheredynamics.com)
  547. ---
  548. ## Support & Contributing
  549. - Report issues: [GitHub Issues](https://github.com/javiarmesto/ALDC-AL-Development-Collection/issues)
  550. - Ask questions: [GitHub Discussions](https://github.com/javiarmesto/ALDC-AL-Development-Collection/discussions)
  551. - See [CONTRIBUTING.md](CONTRIBUTING.md) for contribution guidelines
  552. ---
  553. ## License
  554. MIT — See [LICENSE](LICENSE) for details.
  555. <div align="center">
  556. **Status:** ALDC Core v1.2 COMPLIANT · **Distributions:** Copilot Chat / CLI, Claude Code, Codex · **Root package:** 5.3.0 · **Claude Code:** 5.3.0 · **VS Code Marketplace:** see badge
  557. </div>
  558. ### Independent review
  559. Use **Developer Reviewer** after a direct Developer increment; use the Conductor's
  560. review subagent for orchestrated phases. **Dredd** remains the independent advisory
  561. auditor for an explicit file set, changes or a broader codebase. All three share
  562. provider execution and coverage rules: loading BCQuality is not execution, and
  563. skipping its optional index refresh does not stop path-based review. A partial
  564. review never becomes approval from zero findings. See
  565. [the review contract](docs/templates/review-report-contract.md).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.