Skip to content

Commit 6ad93f1

Browse files
committed
Rewrite README and create docs subpages for maister rebrand
- README rewritten from scratch: concise hero, getting started, standards differentiator, compact workflow table, links to docs/ - docs/workflows.md: detailed phases for all 6 workflow types from actual SKILL.md sources - docs/commands.md: complete command reference from actual command files
1 parent 4a7d169 commit 6ad93f1

File tree

3 files changed

+475
-221
lines changed

3 files changed

+475
-221
lines changed

README.md

Lines changed: 49 additions & 221 deletions
Original file line numberDiff line numberDiff line change
@@ -1,257 +1,85 @@
1-
# AI SDLC Plugin
1+
# Maister
22

3-
> Streamline your software development lifecycle with AI-powered workflows for Claude Code
3+
> Structured, standards-aware development workflows for Claude Code
44
5-
The AI SDLC plugin brings structured, test-driven development workflows to your Claude Code projects. From feature development to bug fixes, performance optimization to security remediation—each task type gets a tailored, guided workflow with automated planning, implementation, and verification.
5+
Maister brings guided workflows to your Claude Code projects. Describe what you want to build, and the plugin handles the rest — from specification through implementation to verification — while enforcing your project's coding standards at every step.
66

7-
## What is AI SDLC Plugin?
7+
## What You Get
88

9-
AI SDLC is a comprehensive development workflow plugin that helps teams:
9+
- **Guided workflows** for features, bug fixes, enhancements, performance, migrations, and research
10+
- **Auto-discovered standards** from your codebase — config files, source patterns, and documentation are analyzed and enforced throughout every workflow
11+
- **Test-driven implementation** with automated planning, incremental verification, and full test suite runs before completion
12+
- **Pause and resume** any workflow — state is preserved across sessions
13+
- **Production readiness checks** including code review, reality assessment, and pragmatic over-engineering detection
1014

11-
- **Plan systematically** with automatic task breakdown and dependency management
12-
- **Implement safely** with test-driven development and continuous standards discovery
13-
- **Verify thoroughly** with automated testing, code review, and production readiness checks
14-
- **Document consistently** with built-in documentation generation and user guides
15-
- **Track transparently** with detailed work logs and verification reports
15+
## Getting Started
1616

17-
Instead of ad-hoc development, you get structured workflows that adapt to your task type—whether you're building a new feature, fixing a bug, optimizing performance, or planning a migration.
17+
**1. Add the marketplace**
1818

19-
## Quick Start
20-
21-
### Installation
22-
23-
1. Install the plugin in your Claude Code environment
24-
2. Initialize the framework in your project:
25-
26-
```bash
27-
/init-sdlc
2819
```
29-
30-
> **Note**: Initialization analyzes your entire codebase to auto-detect coding standards, tech stack, and project conventions. This may take several minutes depending on project size. During this process, Claude may ask permission questions for file access and tool usage—you can approve these to allow the analysis to proceed.
31-
32-
This analyzes your codebase and creates the `.ai-sdlc/` structure with:
33-
- Project documentation (vision, roadmap, tech stack)
34-
- Coding standards (auto-detected from your codebase)
35-
- Task organization folders
36-
37-
### Your First Task
38-
39-
The simplest way to start is with the `/work` command—it automatically classifies your task and routes to the appropriate workflow:
40-
41-
```bash
42-
/work "Add user profile page with avatar upload"
20+
/plugin marketplace add SkillPanel/ai-sdlc
4321
```
4422

45-
The plugin will:
46-
1. Analyze your description
47-
2. Classify the task type (new feature, enhancement, bug fix, etc.)
48-
3. Show you the proposed workflow
49-
4. Ask for confirmation
50-
5. Guide you through each phase
51-
52-
**Alternative**: Use specific workflow commands directly:
23+
**2. Install the plugin**
5324

54-
```bash
55-
/ai-sdlc:development:new "Add user profile page" --type=feature
56-
/ai-sdlc:development:new "Fix login timeout after 5 minutes" --type=bug
57-
/ai-sdlc:development:new "Add sorting to user table" --type=enhancement
5825
```
59-
60-
### Understanding the Output
61-
62-
Each task creates a structured directory in `.ai-sdlc/tasks/[type]/YYYY-MM-DD-task-name/`:
63-
64-
```
65-
2025-11-17-user-profile-page/
66-
├── metadata.yml # Task tracking and status
67-
├── analysis/
68-
│ ├── requirements.md # Gathered requirements
69-
│ └── visuals/ # Design mockups
70-
├── implementation/
71-
│ ├── spec.md # What to build
72-
│ ├── implementation-plan.md # How to build it
73-
│ └── work-log.md # Activity log
74-
├── verification/
75-
│ └── spec-verification.md # Verification results
76-
└── documentation/ # User-facing docs
26+
/plugin install maister@maister-plugins
7727
```
7828

79-
## Key Features
80-
81-
### 5 Adaptive Task Types
82-
83-
Each task type has a specialized workflow optimized for its unique needs:
84-
85-
| Task Type | Use When | Key Features |
86-
|-----------|----------|--------------|
87-
| **New Feature** | Adding completely new capability | 6-7 phase workflow, optional E2E testing, user documentation |
88-
| **Bug Fix** | Fixing defects and errors | TDD Red→Green enforcement, root cause analysis, regression prevention |
89-
| **Enhancement** | Improving existing features | Existing feature analysis, gap detection, backward compatibility |
90-
| **Performance** | Optimizing speed/efficiency | Profiling, bottleneck detection, benchmark validation |
91-
| **Migration** | Moving tech/patterns | Strategy recommendation, rollback planning, dual-run support |
92-
93-
### Intelligent Workflow Features
29+
**3. Initialize your project**
9430

95-
- **Auto-Classification**: `/work` command automatically detects task type from description
96-
- **Execution Modes**: Interactive (pause between phases) or YOLO (continuous execution)
97-
- **Auto-Recovery**: Intelligent failure detection and recovery strategies
98-
- **State Management**: Pause/resume capability with full state preservation
99-
- **Standards Discovery**: Continuous standards checking throughout implementation
100-
- **User-Centric Analysis**: Gap detection, user journey mapping, accessibility checks
101-
102-
### Comprehensive Verification
103-
104-
- **Test-Driven Development**: Write tests first, implement, then verify
105-
- **Full Test Suite**: Run entire project test suite before completion
106-
- **Code Review**: Automated quality, security, and performance analysis
107-
- **Production Readiness**: Deployment readiness verification
108-
- **Reality Checks**: Pragmatic validation that work actually solves the problem
109-
110-
## Core Workflows
111-
112-
### Feature Development
113-
114-
Build completely new capabilities with a comprehensive 6-7 phase workflow:
115-
116-
```bash
117-
/ai-sdlc:development:new "Add two-factor authentication" --type=feature
11831
```
119-
120-
**Phases**: Specification → Planning → Implementation → Verification → E2E Testing (optional) → User Documentation (optional)
121-
122-
**Best for**: New features, adding capabilities that don't exist yet
123-
124-
---
125-
126-
### Enhancements
127-
128-
Improve existing features with backward compatibility verification:
129-
130-
```bash
131-
/ai-sdlc:development:new "Add export to CSV for user reports" --type=enhancement
32+
/maister:init
13233
```
13334

134-
**Phases**: Existing Feature Analysis → Gap Analysis → Specification → Planning → Implementation → Compatibility Verification
35+
This scans your codebase and creates `.maister/` with auto-detected coding standards, project documentation, and task folders. It may take a few minutes on larger projects.
13536

136-
**Best for**: Improving, extending, or enhancing existing features
37+
**4. Start working**
13738

138-
---
139-
140-
### Bug Fixes
141-
142-
Fix defects with mandatory TDD Red→Green discipline:
143-
144-
```bash
145-
/ai-sdlc:development:new "Login timeout after 5 minutes of inactivity" --type=bug
14639
```
147-
148-
**Phases**: Bug Analysis → Fix Implementation (TDD) → Testing & Verification → Documentation
149-
150-
**Best for**: Fixing bugs, errors, crashes with root cause analysis
151-
152-
---
153-
154-
### Performance Optimization
155-
156-
Optimize speed and efficiency with profiling and benchmarking:
157-
158-
```bash
159-
/ai-sdlc:performance:new "Optimize dashboard loading time"
160-
```
161-
162-
**Phases**: Baseline Profiling → Bottleneck Analysis → Implementation with Benchmarking → Performance Verification → Load Testing
163-
164-
**Best for**: Slow responses, high CPU/memory usage, scaling issues
165-
166-
---
167-
168-
### Migrations
169-
170-
Move technologies or patterns with rollback planning:
171-
172-
```bash
173-
/ai-sdlc:migration:new "Migrate from REST to GraphQL API"
174-
```
175-
176-
**Phases**: Current State Analysis → Target State Definition → Migration Strategy → Planning → Implementation → Cutover Verification
177-
178-
**Best for**: Technology upgrades, platform changes, architecture transitions
179-
180-
---
181-
182-
### Research
183-
184-
Investigate technical questions or gather requirements:
185-
186-
```bash
187-
/ai-sdlc:research:new "How does authentication work in the codebase?"
188-
```
189-
190-
**Phases**: Planning → Information Gathering → Analysis & Synthesis → Documentation
191-
192-
**Best for**: Understanding codebase, exploring best practices, requirements gathering
193-
194-
## Commands Quick Reference
195-
196-
### Workflow Commands
197-
198-
| Command | Description |
199-
|---------|-------------|
200-
| `/work [description]` | Auto-classify and route to appropriate workflow |
201-
| `/ai-sdlc:development:new [desc]` | Start development workflow (bug fix, enhancement, or new feature) |
202-
| `/ai-sdlc:performance:new [desc]` | Start performance optimization |
203-
| `/ai-sdlc:migration:new [desc]` | Start migration workflow |
204-
| `/ai-sdlc:research:new [question]` | Start research workflow |
205-
206-
### Resume Commands
207-
208-
All workflows support pause/resume:
209-
210-
```bash
211-
/ai-sdlc:[workflow]:resume [task-path]
40+
/maister:work "Add user profile page with avatar upload"
21241
```
21342

214-
Example: `/ai-sdlc:development:resume .ai-sdlc/tasks/new-features/2025-11-17-user-profile`
215-
216-
### Utility Commands
217-
218-
| Command | Description |
219-
|---------|-------------|
220-
| `/init-sdlc` | Initialize AI SDLC framework |
221-
| `/ai-sdlc:standards-discover` | Auto-discover coding standards |
222-
| `/ai-sdlc:standards:update [path]` | Update project standards |
223-
| `/ai-sdlc:reviews:code [path]` | Automated code quality analysis |
224-
| `/ai-sdlc:reviews:pragmatic [path]` | Check for over-engineering |
225-
| `/ai-sdlc:reviews:spec-audit [spec-path]` | Audit specification completeness |
226-
| `/ai-sdlc:reviews:reality-check [task-path]` | Validate work actually solves problem |
227-
| `/ai-sdlc:reviews:production-readiness [path]` | Pre-deployment verification |
228-
229-
## Design Principles
230-
231-
### Trust Claude to Reason
43+
## How It Works
23244

233-
The plugin provides principles and patterns, not prescriptive implementations. Documentation guides thinking rather than dictating exact steps.
45+
1. You describe a task in natural language
46+
2. The plugin classifies it (feature, bug, enhancement, etc.) and proposes a workflow
47+
3. You confirm, and it guides you through phases: **requirements → spec → plan → implement → verify**
48+
4. At each phase, it asks for your input and decisions (or pass `--yolo` for continuous execution)
49+
5. You get tested, verified code with a detailed work log
23450

235-
### User-Centric Development
51+
All artifacts are saved in `.maister/tasks/` organized by type and date.
23652

237-
Every workflow includes user journey analysis, gap detection, and accessibility checks. Features aren't complete until users can actually discover and use them.
53+
## Supported Workflows
23854

239-
### Test-Driven Approach
55+
| Type | Command | Use When |
56+
|------|---------|----------|
57+
| **Feature** | `/maister:development-new --type=feature` | Adding new capabilities |
58+
| **Bug Fix** | `/maister:development-new --type=bug` | Fixing defects with TDD Red→Green |
59+
| **Enhancement** | `/maister:development-new --type=enhancement` | Improving existing features |
60+
| **Performance** | `/maister:performance-new` | Optimizing speed or resource usage |
61+
| **Migration** | `/maister:migration-new` | Changing technologies or patterns |
62+
| **Research** | `/maister:research-new` | Investigating questions or gathering requirements |
24063

241-
All implementation follows TDD principles: write tests first, implement, then verify. Bug fixes enforce mandatory Red→Green discipline.
64+
Or just use `/maister:work "description"` and let the plugin classify automatically.
24265

243-
### Continuous Standards Discovery
66+
## Standards-Aware Development
24467

245-
Standards from `.ai-sdlc/docs/INDEX.md` are checked throughout implementation, not just at the start. Standards become relevant as work progresses.
68+
This is the key differentiator. Maister doesn't just run workflows — it learns your project's conventions and enforces them:
24669

247-
### Incremental Safety
70+
- **`/maister:init`** scans config files, source code, and documentation to auto-detect your coding standards
71+
- **Continuous checking** — standards are consulted before specification, during planning, and while coding (not just at the start)
72+
- **`/maister:standards-discover`** refreshes standards from your evolving codebase
73+
- **`/maister:standards-update`** lets you add or refine standards manually
24874

249-
Complex changes are broken into small, testable increments with git checkpoints (refactoring) or rollback procedures (migrations).
75+
Standards live in `.maister/docs/standards/` and are indexed in `.maister/docs/INDEX.md`.
25076

251-
### Evidence-Based Verification
77+
## Execution Modes
25278

253-
All findings must reference actual code. Read-only verification reports issues but doesn't fix them—developers should review and apply fixes intentionally.
79+
- **Interactive** (default): pauses between phases for your review and input
80+
- **YOLO** (`--yolo` flag): runs continuously through all phases, asking only critical questions
25481

255-
---
82+
## Learn More
25683

257-
**Made for teams who want structured, test-driven development workflows without sacrificing flexibility.**
84+
- [Workflow Details](docs/workflows.md) — phases, examples, and task structure for each workflow type
85+
- [Full Command Reference](docs/commands.md) — all workflow, review, utility, and quick commands

0 commit comments

Comments
 (0)