MCAF: CI/CD
Trigger On
- adding or changing CI workflows
- defining release flow or rollback policy
- tightening pipeline quality gates
- writing or reviewing AI-assisted pipeline YAML
Value
- produce a concrete project delta: code, docs, config, tests, CI, or review artifact
- reduce ambiguity through explicit planning, verification, and final validation skills
- leave reusable project context so future tasks are faster and safer
Do Not Use For
- feature-level testing with no pipeline or release impact
- general source-control policy without CI/CD changes
Inputs
- the current pipeline and release flow
- real build, test, analyze, and deploy steps
- environment and rollback constraints
Quick Start
- Read the nearest
AGENTS.md and confirm scope and constraints.
- Run this skill's
Workflow through the Ralph Loop until outcomes are acceptable.
- Return the
Required Result Format with concrete artifacts and verification evidence.
Workflow
- Define the target flow first:
- PR validation
- integration-branch gates
- non-production deployment
- production promotion or release
- Keep pipelines reviewable:
- explicit build, test, and analyze steps
- least-privilege secrets and permissions
- rollback or fail-safe strategy
- Treat AI-generated YAML as draft content until it is reviewed and validated.
- For .NET repositories, make the quality gate explicit:
- formatting ownership
- analyzer ownership
- coverage and report generation
- runner model (
VSTest or Microsoft.Testing.Platform)
- Pull only the references that match the current delivery problem.
Deliver
- CI/CD changes that are explicit, reproducible, and reviewable
- release documentation with rollback thinking
- pipeline rules aligned with MCAF verification
Validate
- every stage has a clear purpose and failure signal
- rollback or safe failure is explicit
- secrets and permissions are minimized
- the pipeline matches the repo’s actual verification model
Ralph Loop
Use the Ralph Loop for every task, including docs, architecture, testing, and tooling work.
- Brainstorm first (mandatory):
- analyze current state
- define the problem, target outcome, constraints, and risks
- generate options and think through trade-offs before committing
- capture the recommended direction and open questions
- Plan second (mandatory):
- write a detailed execution plan from the chosen direction
- list final validation skills to run at the end, with order and reason
- Execute one planned step and produce a concrete delta.
- Review the result and capture findings with actionable next fixes.
- Apply fixes in small batches and rerun the relevant checks or review steps.
- Update the plan after each iteration.
- Repeat until outcomes are acceptable or only explicit exceptions remain.
- If a dependency is missing, bootstrap it or return
status: not_applicable with explicit reason and fallback path.
Required Result Format
status: complete | clean | improved | configured | not_applicable | blocked
plan: concise plan and current iteration step
actions_taken: concrete changes made
validation_skills: final skills run, or skipped with reasons
verification: commands, checks, or review evidence summary
remaining: top unresolved items or none
For setup-only requests with no execution, return status: configured and exact next commands.
Load References
- read
references/ci-cd.md first
- for .NET quality gates, use the external
mcaf-dotnet-quality-ci skill from the Managed Code Skills catalog
Example Requests
- "Design CI for this repo."
- "Tighten our deployment gates and rollback story."
- "Review this GitHub Actions YAML before we trust it."
1---2name: mcaf-ci-cd-43description: "Design or refine CI/CD workflows, quality gates, release flow, and safe AI-assisted pipeline authoring. Use when adding or changing build pipelines, release stages, IaC-driven environments, or deployment rollback policy."4compatibility: "Requires repository access; may update CI workflows, pipeline docs, and release guidance."5metadata:6 source: https://skillmd.com/skills/managedcode/mcaf-ci-cd-47---89# MCAF: CI/CD1011## Trigger On1213- adding or changing CI workflows14- defining release flow or rollback policy15- tightening pipeline quality gates16- writing or reviewing AI-assisted pipeline YAML1718## Value1920- produce a concrete project delta: code, docs, config, tests, CI, or review artifact21- reduce ambiguity through explicit planning, verification, and final validation skills22- leave reusable project context so future tasks are faster and safer2324## Do Not Use For2526- feature-level testing with no pipeline or release impact27- general source-control policy without CI/CD changes2829## Inputs3031- the current pipeline and release flow32- real build, test, analyze, and deploy steps33- environment and rollback constraints3435## Quick Start36371. Read the nearest `AGENTS.md` and confirm scope and constraints.382. Run this skill's `Workflow` through the `Ralph Loop` until outcomes are acceptable.393. Return the `Required Result Format` with concrete artifacts and verification evidence.4041## Workflow42431. Define the target flow first:44 - PR validation45 - integration-branch gates46 - non-production deployment47 - production promotion or release482. Keep pipelines reviewable:49 - explicit build, test, and analyze steps50 - least-privilege secrets and permissions51 - rollback or fail-safe strategy523. Treat AI-generated YAML as draft content until it is reviewed and validated.534. For .NET repositories, make the quality gate explicit:54 - formatting ownership55 - analyzer ownership56 - coverage and report generation57 - runner model (`VSTest` or `Microsoft.Testing.Platform`)585. Pull only the references that match the current delivery problem.5960## Deliver6162- CI/CD changes that are explicit, reproducible, and reviewable63- release documentation with rollback thinking64- pipeline rules aligned with MCAF verification6566## Validate6768- every stage has a clear purpose and failure signal69- rollback or safe failure is explicit70- secrets and permissions are minimized71- the pipeline matches the repo’s actual verification model7273## Ralph Loop7475Use the Ralph Loop for every task, including docs, architecture, testing, and tooling work.76771. Brainstorm first (mandatory):78 - analyze current state79 - define the problem, target outcome, constraints, and risks80 - generate options and think through trade-offs before committing81 - capture the recommended direction and open questions822. Plan second (mandatory):83 - write a detailed execution plan from the chosen direction84 - list final validation skills to run at the end, with order and reason853. Execute one planned step and produce a concrete delta.864. Review the result and capture findings with actionable next fixes.875. Apply fixes in small batches and rerun the relevant checks or review steps.886. Update the plan after each iteration.897. Repeat until outcomes are acceptable or only explicit exceptions remain.908. If a dependency is missing, bootstrap it or return `status: not_applicable` with explicit reason and fallback path.9192### Required Result Format9394- `status`: `complete` | `clean` | `improved` | `configured` | `not_applicable` | `blocked`95- `plan`: concise plan and current iteration step96- `actions_taken`: concrete changes made97- `validation_skills`: final skills run, or skipped with reasons98- `verification`: commands, checks, or review evidence summary99- `remaining`: top unresolved items or `none`100101For setup-only requests with no execution, return `status: configured` and exact next commands.102103## Load References104105- read `references/ci-cd.md` first106- for .NET quality gates, use the external `mcaf-dotnet-quality-ci` skill from the [Managed Code Skills catalog](https://skills.managed-code.com/)107108## Example Requests109110- "Design CI for this repo."111- "Tighten our deployment gates and rollback story."112- "Review this GitHub Actions YAML before we trust it."