By wilsonkichoi
AI-Assisted Software Development Workflow — 5-phase pipeline (Research, Specification, Task Breakdown, Execution, Verification) with role-based agents, decision records, and task review files.
Backend engineer for API implementation, data layer, and service logic. Implements API endpoints, service logic, and database operations.
Code reviewer for quality review of completed work. Reviews correctness, security, maintainability, and patterns against spec and project standards.
Data engineer for schema implementation, migrations, and data integrity. Implements database migrations, data models, and data pipelines.
DevOps engineer for infrastructure, CI/CD, and deployment. Implements infrastructure-as-code, pipeline setup, and deployment configuration.
Domain expert ensuring technical designs match business reality. Reviews systems with complex business rules or industry-specific logic.
Use to run the full execute-review-fix-verify pipeline automatically with clean context between steps. Orchestrates sequential subagents for unattended execution. Use when someone says 'auto wave', 'auto milestone', 'run full pipeline', 'automated execution', 'hands-free', 'run everything', 'unattended wave', or 'auto run'.
Use for Phase 4 of the development workflow: task execution. Implements one task or an entire wave with role-matched expertise, branch isolation, task review files, and debugging framework. Use when someone says 'execute task', 'phase 4', 'implement next task', 'start execution', 'pick up next task', 'execute wave', or 'run wave N'.
Use when someone asks to initialize a new project, scaffold a project, set up the development workflow, or create project structure. Also trigger on 'start new project', 'workflow setup', 'project scaffold', or 'init from template'. Even if they just say 'set up a new project', this skill is likely what they want.
Use for Phase 3 of the development workflow: task breakdown. Decomposes milestones into atomic tasks grouped into parallel waves with role assignments. Use when someone says 'create plan', 'phase 3', 'task breakdown', 'decompose tasks', 'break down the spec', or after Phase 2 spec is approved.
Use for Phase 1 of the development workflow: AI deep research. Ingests human research materials from workflow/research/manual/ and performs web search, codebase analysis, and cross-referencing to produce a comprehensive research synthesis. Use when someone says 'start research', 'phase 1', 'research phase', 'deep research', or has placed materials in workflow/research/manual/.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A structured, 5-phase framework for building software with AI coding agents. Gives LLMs clear specifications, atomic tasks, and role-matched expertise — so you get predictable, high-quality output instead of hoping the AI figures it out.
Note: I created this project to better understand agentic software development methodology. There may be a bunch of stupid stuff here. For real projects, check out Superpowers, GSD, Spec Kit, EPCC, oh-my-claudecode, Compound Engineering, Agent Skills and incorporate whatever works for you.
Most people use AI coding tools by throwing a vague prompt at an LLM and hoping for the best. This works for small tasks but falls apart for anything real:
This workflow breaks AI-assisted development into 5 phases with human gates between each:
Phase 1 Phase 2 Phase 3 Phase 4 Phase 5
Research --> Specification --> Task Breakdown --> Execution --> Verification
[GATE] [GATE] [GATE] [GATE]
Phase 1: Research — Human gathers materials, AI performs deep research to fill gaps and cross-reference findings.
Phase 2: Specification — AI generates architecture specs (SPEC.md) with milestones, schemas, API contracts, and diagrams. No code. Optional multi-role review (security, domain, product).
Phase 3: Task Breakdown — Spec is decomposed into atomic tasks grouped into parallel waves, with role assignments and test separation.
Phase 4: Execution — One task per session, role-matched agents, automated verification, human review via task review files, PR on approval.
Phase 5: Verification — Per-milestone end-to-end testing and retrospective.
Each phase produces persistent documents (SPEC.md, PLAN.md, PROGRESS.md, decision records, task reviews) that survive across sessions and prevent context rot.
| Component | Count | Description |
|---|---|---|
| Skills | 8 | init-project, research, spec, plan, execute, review, verify, auto |
| Agents | 13 | Software Architect, Backend Engineer, QA Engineer, Security Reviewer, and more |
# Add the marketplace
/plugin marketplace add wilsonkichoi/agentic_development_workflow
# Install the plugin — select "Install for you, in this repo only (local scope)"
/plugin install agentic-dev@wilsonkichoi-agentic-dev
When prompted for install scope, choose "Install for you, in this repo only" (local scope). This keeps the plugin scoped to your project rather than applying globally, which is recommended while the workflow is still evolving.
After installation:
/agentic-dev:research, /agentic-dev:spec, etc./agents (e.g., agentic-dev:backend-engineer)copilot plugin install wilsonkichoi/agentic_development_workflow
After installation:
/init-project, /research, /spec, /plan, /execute, /review, /verify, /auto/agent (e.g., backend-engineer, qa-engineer)~/.copilot/state/installed-plugins/Note: Copilot CLI does not support per-project plugin scoping. Installed plugins are always global — there is no
--scopeflag or disable/enable per workspace (unlike Gemini CLI). If you need project-level control, use Claude Code's local scope install or Gemini CLI's workspace scope instead.
To avoid polluting the global context, install the extension globally but keep it disabled by default. Enable it strictly on a per-workspace basis:
# 1. Install the extension from GitHub
gemini extensions install https://github.com/wilsonkichoi/agentic_development_workflow
# 2. Disable the extension globally (user scope) so it doesn't pollute all your projects
gemini extensions disable agentic-development-workflow --scope user
# 3. Navigate into specific projects and enable it locally (workspace scope)
cd /path/to/your-project
gemini extensions enable agentic-development-workflow --scope workspace
npx claudepluginhub wilsonkichoi/agentic_development_workflow --plugin agentic-devResume management with setup, import, generate, tailor, score, match, review, verify, ingest, research, qualify, cover-letter, and apply skills. Anti-fabrication enforcement, company research, strategic qualification, multi-persona review, and end-to-end application pipeline.
GSD Core is a meta-prompting, context engineering, and spec-driven development system for AI coding agents.
Plan and autonomously build a software task end-to-end. Recons the codebase, applies preloaded memory, decomposes into the right number of phases, gets one confirmation, then prepares a single ready-to-paste /goal command — one paste between you and done — that drives execution to completion with built-in retry, fix-spec recovery, and per-phase memory writeback. Works on Claude Code and Codex.
Corca Workflow Framework — consolidated hooks and skill orchestration for structured development sessions
Persona-driven AI development team: orchestrator, team agents, review agents, skills, slash commands, and advisory hooks for Claude Code
A curated set of skills for each stage of development — propose, spec, design, plan, implement, ship.
Specification-driven development workflow: specify → plan → tasks → implement