ahtishamshahzad/agent_dev_flow/ci-cd

Trust tier 3
  • community
  • ext:[{"type":"hook","description":"Pre/post lifecycle hooks"}]
Source check pending

Checking that the SKILL.md file is still in the repository.

Use to design the CI/CD pipeline generated from the project's selected applications and test selection — jobs like install, lint, typecheck, unit, API, Playwright, mobile, build, security checks, deployment. Only generate jobs for applications that exist; deploys are approval-gated. Does not scaffold infrastructure.

Install

npx vskill@latest install ahtishamshahzad/agent_dev_flow/ci-cd

Semi-extensibleMentions customization, but needs skill-specific knowledge. Read the extensible skills standard

SKILL.md

CI/CD

Purpose

Design a CI/CD pipeline that reflects what the project actually is — its selected applications and chosen test types — running the right checks in the right order, gating merges and (with approval) deploys. Platform wiring for GitHub is github-actions.

When to Use

  • When establishing or revising the pipeline for a project.
  • After applications, tests, and deployment target are selected.
  • Not to scaffold infrastructure, or to add jobs for applications that don't exist.

Inputs

  • Selected applications (../../application-selection) — the pipeline is generated from these.
  • Test selection (../../testing/testing-selection), repo/monorepo shape (repository-strategy, monorepo-selection), deployment target (deployment-selection).

Discovery Questions

  • Which applications exist (backend, web, dashboard, mobile, shared packages)? The pipeline covers exactly these.
  • Which test types were selected per app (unit, API/Supertest, Playwright, Maestro)?
  • Is this a monorepo where affected-only runs and caching apply (turborepo-foundation)?
  • What is gated at merge vs deploy, and does deploy require approval?

Responsibilities

  • Generate jobs from the selected applications + test selection, choosing from: install, lint, typecheck, unit tests, API tests, Playwright (web/dashboard E2E), mobile tests (Maestro), build, security checks, deployment — including only those an existing application needs.
  • Never create jobs for applications that don't exist: no Playwright job without a web app, no mobile job without a mobile app, no deploy job without a deployable — absence of an app means absence of its jobs.
  • Order for fast feedback: cheap/fast checks (lint, typecheck, unit) before slow ones (E2E, build); fail fast; parallelize independent jobs.
  • Use frozen, cached installs (pnpm --frozen-lockfile/npm ci); in monorepos, affected-only execution + caching (turborepo-foundation).
  • Wire security checks (dependency/secret scanning — ../../security/dependency-security, secrets-audit) as pipeline jobs.
  • Set gates: required checks block merge; deployment jobs are approval-gated and run only after tests + security pass (../../backend/backend-deployment, production-readiness, ../../release-planning).
  • Provide environment/config validation and post-deploy smoke as pipeline steps (environment-management, ../../testing/smoke-testing).

Required Workflow

  1. Read the selected applications + test selection + repo shape.
  2. Derive the job set per app (only for apps that exist).
  3. Order jobs fast→slow; parallelize; frozen/affected installs.
  4. Add security-check jobs.
  5. Set merge gates; make deploy approval-gated with post-deploy smoke.
  6. Hand platform wiring to github-actions.

Decision Rules

  • The pipeline is a function of the selected applications — derive it, don't template a fixed pipeline onto every project.
  • No job for a nonexistent application, ever.
  • Fast checks gate before expensive ones; a failing lint shouldn't wait behind E2E.
  • Deploys never run unapproved or on red tests/security (../../release-planning gate 7).

Rules

  • Jobs map to real applications and selected test types.
  • Deployment is approval-gated; no auto-deploy without explicit policy.
  • No infrastructure scaffolding — pipeline design only.

Anti-Patterns

  • A fixed one-size pipeline with jobs for apps the project lacks.
  • Playwright/mobile jobs with no corresponding app.
  • Slow E2E running before cheap lint/typecheck.
  • npm install (unpinned) in CI instead of npm ci.
  • Auto-deploying on green without approval or post-deploy smoke.

Validation Checklist

  • [ ] Job set derived from selected applications + test selection.
  • [ ] No jobs for nonexistent applications.
  • [ ] Fast→slow ordering; parallel; frozen/affected installs.
  • [ ] Security-check jobs included.
  • [ ] Merge gates set; deploy approval-gated with post-deploy smoke.
  • [ ] Platform wiring handed to github-actions.

Definition of Done

A pipeline design generated from the project's real applications and selected tests — correctly ordered, frozen/affected installs, security checks, merge gates, and approval-gated deploys with post-deploy smoke — with no jobs for applications that don't exist.

Related Skills

github-actions, ../../testing/testing-selection, ../../application-selection, repository-strategy, monorepo-selection, turborepo-foundation, deployment-selection, production-readiness, ../../security/dependency-security, ../../release-planning.

Related Knowledge

../../../knowledge/ (selected apps, deploy target).

Related References

../../../references/devops/ (pipeline patterns, when populated).

Context Loading Guidance

  • Requires: selected applications, test selection, repo shape, deploy target.
  • Does not require: infrastructure provisioning, app feature code.
  • May load: github-actions, deployment-selection.
  • Stop when: the app-derived job set + gates are designed.

Token Efficiency Guidance

The application × job matrix is the artifact — it directly encodes "only jobs for apps that exist." Build it, then wire the platform.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Extension points

  • hookPre/post lifecycle hooks

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/ahtishamshahzad/agent_dev_flow/ci-cd/badge)](https://verified-skill.com/skills/ahtishamshahzad/agent_dev_flow/ci-cd)

Something wrong with this skill or its page? Tell us.