dotnet-mcaf
Adopt MCAF governance in a .NET repository with the right AGENTS.md layout, repo-native docs, skill installation, verification rules, and non-trivial task workflow. Use when bootstrapping or updating MCAF alongside the dotnet-skills catalog.
Install
npx skills add https://github.com/Postpartum-genushyacinthus29/dotnet-skills/tree/main/skills/dotnet-mcaf
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install postpartum-genushyacinthus29-dotnet-skills@llmmart
git clone https://github.com/Postpartum-genushyacinthus29/dotnet-skills.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole postpartum-genushyacinthus29/dotnet-skills collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
MCAF Adoption
Trigger On
- bootstrapping MCAF in a new or existing repository that also contains
.NETwork - updating root or project-local
AGENTS.mdfiles to follow a durable repo workflow - deciding which MCAF governance skills and
dotnet-*implementation skills to install together - organizing repo-native docs for architecture, features, ADRs, testing, development, and operations
Workflow
Start from the canonical bootstrap surface:
- tutorial:
https://mcaf.managed-code.com/tutorial - concepts:
https://mcaf.managed-code.com/ - public MCAF skills:
https://mcaf.managed-code.com/skills
- tutorial:
Place root
AGENTS.mdat the repository or solution root.Add project-local
AGENTS.mdonly when the solution has multiple projects with genuinely different local rules.Install MCAF governance skills (
dotnet-mcaf-*) for process areas anddotnet-*implementation skills for framework work. Checkreferences/skill-map.mdfor overlap before adding duplicate surfaces.Route to the narrowest MCAF skill once the governance concern is clear:
Concern Skill Delivery workflow and feedback loops dotnet-mcaf-agile-deliveryDeveloper onboarding and local inner loop dotnet-mcaf-devexDurable docs structure and source-of-truth placement dotnet-mcaf-documentationExecutable feature behaviour docs dotnet-mcaf-feature-specHuman review for large AI-generated drops dotnet-mcaf-human-review-planningML/AI product delivery process dotnet-mcaf-ml-ai-deliveryExplicit quality attributes and trade-offs dotnet-mcaf-nfrBranch, merge, and release hygiene dotnet-mcaf-source-controlDesign-system, accessibility, front-end direction dotnet-mcaf-ui-uxScaffold repo-native documentation:
docs/ ├── Architecture.md ├── Features/ ├── ADR/ ├── Testing/ ├── Development/ └── Operations/Encode the non-trivial task flow in
AGENTS.md:<slug>.brainstorm.mdthen<slug>.plan.mdthen implementation and validation.Treat verification as part of done: tests, analyzers, formatters, coverage, and any architecture or security gates the repo configured.
flowchart LR
A["Adopt MCAF"] --> B["Root AGENTS.md"]
B --> C{"Multi-project?"}
C -->|Yes| D["Project-local AGENTS.md"]
C -->|No| E["Root policy only"]
B --> F["Install mcaf-* governance skills"]
B --> G["Install dotnet-* implementation skills"]
D --> H["Document boundaries and commands"]
E --> H
F --> I["Repo-native docs scaffolds"]
G --> J[".NET implementation guidance"]
H --> K["Run full quality pass"]
I --> K
J --> K
Deliver
- repository-ready MCAF adoption with clear root and local
AGENTS.mdresponsibilities - correct split between
mcaf-*governance anddotnet-*implementation skills - repo-native docs and verification expectations instead of chat-only instructions
Validate
- root
AGENTS.mdexists at the repository or solution root - project-local
AGENTS.mdfiles exist only where genuinely needed - repo documents exact build, test, format, analyze, and coverage commands
- durable docs exist for architecture and behavior, not only inline comments
- non-trivial work follows the brainstorm-to-plan flow before implementation
- the full quality pass is part of done, not only a narrow happy-path test run
References
- references/adoption.md - canonical MCAF entry points, bootstrap rules, and the local-mirror boundary between governance and implementation skills
- references/skill-map.md - MCAF catalog map with overlap-vs-new split for precise routing
Files (dotnet-skills)
-
references
-
adoption.md 4.6 KB
# MCAF Adoption Alongside dotnet-skills Use this reference when a repository wants to adopt the Managed Code Coding AI Framework and also uses the dotnet-skills catalog for `.NET` implementation work instead of keeping AI workflow rules scattered across chat threads, tribal knowledge, or ad hoc prompts. ## Canonical Sources - Concepts: `https://mcaf.managed-code.com/` - Tutorial: `https://mcaf.managed-code.com/tutorial` - Skills catalog: `https://mcaf.managed-code.com/skills` - Source repository: `https://github.com/managedcode/MCAF` The concepts page defines the framework. The tutorial is the canonical bootstrap flow. ## What MCAF Adds To A Repository From the official concepts and tutorial pages, the core MCAF shape is: 1. durable engineering context stays in the repository 2. agents work from `AGENTS.md`, repo docs, and installed skills 3. verification is enforced through tests and analyzers 4. guidance stays small, explicit, versioned, and repo-native For repositories that also contain `.NET` code, the official tutorial makes one boundary explicit: - install MCAF governance skills from the MCAF catalog - install implementation-focused `dotnet-*` skills from `https://skills.managed-code.com/` That means MCAF is the framework layer, while this repository remains the `.NET` execution layer. The local `dotnet-skills` catalog now mirrors the net-new MCAF governance surfaces as installable `dotnet-mcaf-*` skills. Use the local mirrors first for the transferred surfaces, and fall back to the upstream MCAF catalog when a governance skill has not been mirrored here yet. ## Bootstrap Rules That Matter Alongside dotnet-skills Use the tutorial as the single canonical install surface. Required bootstrap artifacts: - root `AGENTS.md` at the repo or solution root - optional `CLAUDE.md` wrapper when the repo uses Claude Code - local `dotnet-mcaf*` governance skills under the native agent skill directory - `dotnet-*` implementation skills from this catalog when the repo contains `.NET` code For multi-project solutions: - keep one root `AGENTS.md` - add one local `AGENTS.md` per project or module root when local rules differ - local files may tighten root policy but must not silently weaken it ## Repo-Native Documentation Shape The official MCAF concepts page recommends durable docs under: - `docs/Architecture.md` - `docs/Features/` - `docs/ADR/` - `docs/Testing/` - `docs/Development/` - `docs/Operations/` For `.NET` teams, that means build and validation knowledge should not stay only in CI YAML or chat history. The repo should explicitly document: - exact `dotnet build` commands - exact `dotnet test` commands - formatter and analyzer commands - coverage commands and thresholds when used - architecture and behavior that tests must prove ## Task Workflow Expectations The official MCAF guidance expects non-trivial work to follow: 1. root-level `<slug>.brainstorm.md` 2. root-level `<slug>.plan.md` 3. implementation 4. validation against the repo-defined quality pass Do not compress architecture choice, test strategy, and execution into one improvised edit cycle when the task is non-trivial. ## Verification Expectations Official MCAF verification rules explicitly favor: - TDD for new behavior and bug fixes - user-visible or caller-visible scenario coverage - integration and end-to-end checks when behavior crosses boundaries - static analysis as part of done - full relevant suites green before completion For `.NET` repos this means “tests passed” is insufficient if the repo also requires: - analyzers - `dotnet format` - coverage - architecture tests - security gates ## MCAF Skill Map That Usually Matters Alongside dotnet-skills The local `managedcode/MCAF` catalog currently exposes 18 separate `mcaf-*` skills. In practice that means teams should route to a narrow governance skill, not cite "MCAF" generically. Recommended split: - use MCAF skills for governance, docs, planning, maintainability, testing policy, CI/CD policy, observability, security baseline, and delivery process - use `dotnet-*` skills from this catalog for framework-specific implementation and validation The most common `.NET` adoption map is in `references/skill-map.md`. ## Practical Routing For .NET Teams When the ask is: - "set up repo rules, AGENTS, docs, and delivery workflow" -> start with `dotnet-mcaf` - "implement or fix actual ASP.NET/Orleans/EF/Agent Framework code" -> route from `dotnet-mcaf` into the narrowest `dotnet-*` skill - "tighten verification and CI rules" -> combine `dotnet-mcaf` with the appropriate testing or quality skill - "choose which local MCAF mirror should own governance work" -> route through the grouped map in `references/skill-map.md` -
skill-map.md 6.7 KB
# Current MCAF Skill Map This map is sourced from the current local `managedcode/MCAF` catalog under `skills/`. Use it when a `.NET` repository wants MCAF but the task is specific enough that "use MCAF" is too vague to be useful. ## Governance And Delivery Flow | MCAF skill | Use it for | |---|---| | `mcaf-solution-governance` | root and local `AGENTS.md`, rule precedence, topology, maintainability-limit placement, and solution-wide agent policy | | `mcaf-agile-delivery` | backlog quality, planning flow, ceremonies, and turning delivery feedback into durable process changes | | `mcaf-source-control` | branch naming, merge strategy, commit hygiene, release-policy guardrails, and secrets-in-git discipline | | `mcaf-human-review-planning` | large AI-generated changes that need a practical human review sequence instead of a flat file-by-file read | ## Local Mirrors In This Catalog The following net-new MCAF surfaces are now mirrored locally in `dotnet-skills`: | Canonical MCAF skill | Local mirror in this catalog | |---|---| | `mcaf-agile-delivery` | `dotnet-mcaf-agile-delivery` | | `mcaf-devex` | `dotnet-mcaf-devex` | | `mcaf-documentation` | `dotnet-mcaf-documentation` | | `mcaf-feature-spec` | `dotnet-mcaf-feature-spec` | | `mcaf-human-review-planning` | `dotnet-mcaf-human-review-planning` | | `mcaf-ml-ai-delivery` | `dotnet-mcaf-ml-ai-delivery` | | `mcaf-nfr` | `dotnet-mcaf-nfr` | | `mcaf-source-control` | `dotnet-mcaf-source-control` | | `mcaf-ui-ux` | `dotnet-mcaf-ui-ux` | ## Docs And Architecture | MCAF skill | Use it for | |---|---| | `mcaf-architecture-overview` | creating or updating `docs/Architecture.md` as the global system map | | `mcaf-feature-spec` | feature behavior specs under `docs/Features/` | | `mcaf-adr-writing` | ADRs under `docs/ADR/` for technical decisions and trade-offs | | `mcaf-documentation` | durable engineering docs structure, navigation, source-of-truth placement, and writing quality | | `mcaf-nfr` | explicit non-functional requirements such as reliability, accessibility, maintainability, scalability, and compliance | ## Quality, Testing, And Review | MCAF skill | Use it for | |---|---| | `mcaf-testing` | repository-aligned automated tests, verification flows, and test strategy updates | | `mcaf-code-review` | PR scope, review checklists, reviewer expectations, and merge hygiene | | `mcaf-solid-maintainability` | SOLID, SRP, cohesion, splitting large files or classes, and maintainability-limit enforcement | | `mcaf-security-baseline` | secure defaults, secrets handling, review checkpoints, and baseline security guardrails | | `mcaf-observability` | logs, metrics, traces, diagnostics, alerts, and runtime visibility policy | ## Tooling, Ops, And Product Surfaces | MCAF skill | Use it for | |---|---| | `mcaf-ci-cd` | CI/CD pipelines, quality gates, release flow, deployment stages, and rollback policy | | `mcaf-devex` | onboarding, local inner loop, reproducible setup, and developer workflow quality | | `mcaf-ui-ux` | design-system, accessibility, front-end technology selection, and design-to-dev collaboration | | `mcaf-ml-ai-delivery` | ML or AI product delivery, experimentation, responsible-AI workflow, and model/inference delivery planning | ## Practical .NET Routing Start with `dotnet-mcaf` when the ask is "adopt MCAF" or "make this repo follow MCAF". Then route: - repo bootstrap and root/local `AGENTS.md` work -> `dotnet-mcaf` - delivery workflow -> `dotnet-mcaf-agile-delivery` - developer onboarding and local loop -> `dotnet-mcaf-devex` - docs bootstrap -> `dotnet-mcaf-documentation` - feature behaviour docs -> `dotnet-mcaf-feature-spec` - large generated-drop review sequencing -> `dotnet-mcaf-human-review-planning` - ML or AI delivery policy -> `dotnet-mcaf-ml-ai-delivery` - source-control policy -> `dotnet-mcaf-source-control` - explicit quality attributes -> `dotnet-mcaf-nfr` - UI/UX and accessibility direction -> `dotnet-mcaf-ui-ux` - overlapping architecture, testing, CI, security, observability, and maintainability areas -> keep the boundary guidance below and route into the existing `dotnet-*` implementation skills as needed After governance routing is clear, switch to the matching `dotnet-*` skill for real framework or code changes. ## Overlap Versus Net-New Surface Some MCAF skills overlap conceptually with areas that already exist in `dotnet-skills`, but they operate at a different layer. ### Conceptual overlap with existing `dotnet-*` skills | MCAF skill | Closest current `dotnet-skills` surface | Boundary | |---|---|---| | `mcaf-architecture-overview` | `dotnet-architecture` | MCAF defines repo architecture docs and decision shape; `dotnet-architecture` covers actual .NET solution structure and technical design | | `mcaf-code-review` | `dotnet-code-review` | MCAF defines review process and merge hygiene; `dotnet-code-review` reviews .NET code changes for bugs and regressions | | `mcaf-testing` | `dotnet-quality-ci`, test-framework skills such as `dotnet-xunit`, `dotnet-nunit`, `dotnet-mstest`, `dotnet-tunit` | MCAF defines verification policy; `dotnet-*` skills define concrete .NET test and CI implementation | | `mcaf-ci-cd` | `dotnet-quality-ci`, `dotnet-project-setup` | MCAF defines release-flow and governance policy; `dotnet-*` skills wire concrete .NET pipeline commands and quality gates | | `mcaf-solution-governance` | `dotnet-project-setup` | MCAF defines root/local `AGENTS.md`, rule precedence, and repo policy; `dotnet-project-setup` defines solution and project structure | | `mcaf-solid-maintainability` | `dotnet-complexity`, analyzer skills | MCAF defines maintainability expectations; `dotnet-*` skills implement concrete metrics, analyzers, and refactoring mechanics | | `mcaf-security-baseline` | `dotnet-codeql`, analyzer and platform security skills | MCAF defines baseline security process; `dotnet-*` skills cover concrete .NET tooling and framework-specific security work | | `mcaf-observability` | platform/runtime skills such as `dotnet-aspnet-core`, `dotnet-worker-services`, `dotnet-aspire`, `dotnet-orleans` | MCAF defines telemetry policy; `dotnet-*` skills implement logging, tracing, metrics, and diagnostics per framework | ### Surfaces that are effectively new relative to this catalog These MCAF skills did not have a close one-to-one equivalent in the original `dotnet-skills` catalog and are now mirrored locally as: - `dotnet-mcaf-agile-delivery` - `dotnet-mcaf-devex` - `dotnet-mcaf-documentation` - `dotnet-mcaf-feature-spec` - `dotnet-mcaf-human-review-planning` - `dotnet-mcaf-ml-ai-delivery` - `dotnet-mcaf-nfr` - `dotnet-mcaf-source-control` - `dotnet-mcaf-ui-ux` Treat those as genuinely additive. They extend repo workflow and delivery governance rather than duplicating .NET implementation guidance.
-
-
SKILL.md 4.1 KB
--- name: dotnet-mcaf version: "1.2.1" category: "Core" description: "Adopt MCAF governance in a .NET repository with the right AGENTS.md layout, repo-native docs, skill installation, verification rules, and non-trivial task workflow. Use when bootstrapping or updating MCAF alongside the dotnet-skills catalog." compatibility: "Best for repositories that want MCAF governance and also use dotnet-skills for actual .NET implementation work." --- # MCAF Adoption ## Trigger On - bootstrapping MCAF in a new or existing repository that also contains `.NET` work - updating root or project-local `AGENTS.md` files to follow a durable repo workflow - deciding which MCAF governance skills and `dotnet-*` implementation skills to install together - organizing repo-native docs for architecture, features, ADRs, testing, development, and operations ## Workflow 1. Start from the canonical bootstrap surface: - tutorial: `https://mcaf.managed-code.com/tutorial` - concepts: `https://mcaf.managed-code.com/` - public MCAF skills: `https://mcaf.managed-code.com/skills` 2. Place root `AGENTS.md` at the repository or solution root. 3. Add project-local `AGENTS.md` only when the solution has multiple projects with genuinely different local rules. 4. Install MCAF governance skills (`dotnet-mcaf-*`) for process areas and `dotnet-*` implementation skills for framework work. Check `references/skill-map.md` for overlap before adding duplicate surfaces. 5. Route to the narrowest MCAF skill once the governance concern is clear: | Concern | Skill | |---------|-------| | Delivery workflow and feedback loops | `dotnet-mcaf-agile-delivery` | | Developer onboarding and local inner loop | `dotnet-mcaf-devex` | | Durable docs structure and source-of-truth placement | `dotnet-mcaf-documentation` | | Executable feature behaviour docs | `dotnet-mcaf-feature-spec` | | Human review for large AI-generated drops | `dotnet-mcaf-human-review-planning` | | ML/AI product delivery process | `dotnet-mcaf-ml-ai-delivery` | | Explicit quality attributes and trade-offs | `dotnet-mcaf-nfr` | | Branch, merge, and release hygiene | `dotnet-mcaf-source-control` | | Design-system, accessibility, front-end direction | `dotnet-mcaf-ui-ux` | 6. Scaffold repo-native documentation: ``` docs/ ├── Architecture.md ├── Features/ ├── ADR/ ├── Testing/ ├── Development/ └── Operations/ ``` 7. Encode the non-trivial task flow in `AGENTS.md`: `<slug>.brainstorm.md` then `<slug>.plan.md` then implementation and validation. 8. Treat verification as part of done: tests, analyzers, formatters, coverage, and any architecture or security gates the repo configured. ```mermaid flowchart LR A["Adopt MCAF"] --> B["Root AGENTS.md"] B --> C{"Multi-project?"} C -->|Yes| D["Project-local AGENTS.md"] C -->|No| E["Root policy only"] B --> F["Install mcaf-* governance skills"] B --> G["Install dotnet-* implementation skills"] D --> H["Document boundaries and commands"] E --> H F --> I["Repo-native docs scaffolds"] G --> J[".NET implementation guidance"] H --> K["Run full quality pass"] I --> K J --> K ``` ## Deliver - repository-ready MCAF adoption with clear root and local `AGENTS.md` responsibilities - correct split between `mcaf-*` governance and `dotnet-*` implementation skills - repo-native docs and verification expectations instead of chat-only instructions ## Validate - root `AGENTS.md` exists at the repository or solution root - project-local `AGENTS.md` files exist only where genuinely needed - repo documents exact build, test, format, analyze, and coverage commands - durable docs exist for architecture and behavior, not only inline comments - non-trivial work follows the brainstorm-to-plan flow before implementation - the full quality pass is part of done, not only a narrow happy-path test run ## References - references/adoption.md - canonical MCAF entry points, bootstrap rules, and the local-mirror boundary between governance and implementation skills - references/skill-map.md - MCAF catalog map with overlap-vs-new split for precise routing
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.