red64-cli 0.3.0 → 0.6.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +194 -338
- package/dist/cli/parseArgs.d.ts.map +1 -1
- package/dist/cli/parseArgs.js +5 -13
- package/dist/cli/parseArgs.js.map +1 -1
- package/dist/components/init/types.d.ts +0 -2
- package/dist/components/init/types.d.ts.map +1 -1
- package/dist/components/screens/HelpScreen.d.ts.map +1 -1
- package/dist/components/screens/HelpScreen.js +0 -2
- package/dist/components/screens/HelpScreen.js.map +1 -1
- package/dist/components/screens/InitScreen.d.ts.map +1 -1
- package/dist/components/screens/InitScreen.js +5 -8
- package/dist/components/screens/InitScreen.js.map +1 -1
- package/dist/components/screens/StartScreen.d.ts.map +1 -1
- package/dist/components/screens/StartScreen.js +29 -8
- package/dist/components/screens/StartScreen.js.map +1 -1
- package/dist/components/screens/StatusScreen.d.ts.map +1 -1
- package/dist/components/screens/StatusScreen.js +16 -1
- package/dist/components/screens/StatusScreen.js.map +1 -1
- package/dist/services/AgentInvoker.d.ts.map +1 -1
- package/dist/services/AgentInvoker.js +76 -37
- package/dist/services/AgentInvoker.js.map +1 -1
- package/dist/services/ClaudeErrorDetector.d.ts +1 -1
- package/dist/services/ClaudeErrorDetector.d.ts.map +1 -1
- package/dist/services/ClaudeErrorDetector.js +1 -0
- package/dist/services/ClaudeErrorDetector.js.map +1 -1
- package/dist/services/ClaudeHealthCheck.d.ts +7 -0
- package/dist/services/ClaudeHealthCheck.d.ts.map +1 -1
- package/dist/services/ClaudeHealthCheck.js +76 -12
- package/dist/services/ClaudeHealthCheck.js.map +1 -1
- package/dist/services/ConfigService.d.ts +1 -0
- package/dist/services/ConfigService.d.ts.map +1 -1
- package/dist/services/ConfigService.js.map +1 -1
- package/dist/services/DockerRunner.js +1 -1
- package/dist/services/DockerRunner.js.map +1 -1
- package/dist/services/PhaseExecutor.d.ts.map +1 -1
- package/dist/services/PhaseExecutor.js +2 -1
- package/dist/services/PhaseExecutor.js.map +1 -1
- package/dist/services/TaskRunner.d.ts.map +1 -1
- package/dist/services/TaskRunner.js +2 -1
- package/dist/services/TaskRunner.js.map +1 -1
- package/dist/services/index.d.ts +1 -1
- package/dist/services/index.d.ts.map +1 -1
- package/dist/services/index.js +1 -1
- package/dist/services/index.js.map +1 -1
- package/dist/types/index.d.ts +4 -3
- package/dist/types/index.d.ts.map +1 -1
- package/dist/types/index.js.map +1 -1
- package/framework/stacks/c/code-quality.md +326 -0
- package/framework/stacks/c/coding-style.md +347 -0
- package/framework/stacks/c/conventions.md +513 -0
- package/framework/stacks/c/error-handling.md +350 -0
- package/framework/stacks/c/feedback.md +158 -0
- package/framework/stacks/c/memory-safety.md +408 -0
- package/framework/stacks/c/tech.md +122 -0
- package/framework/stacks/c/testing.md +472 -0
- package/framework/stacks/cpp/code-quality.md +282 -0
- package/framework/stacks/cpp/coding-style.md +363 -0
- package/framework/stacks/cpp/conventions.md +420 -0
- package/framework/stacks/cpp/error-handling.md +264 -0
- package/framework/stacks/cpp/feedback.md +104 -0
- package/framework/stacks/cpp/memory-safety.md +351 -0
- package/framework/stacks/cpp/tech.md +160 -0
- package/framework/stacks/cpp/testing.md +323 -0
- package/framework/stacks/java/code-quality.md +357 -0
- package/framework/stacks/java/coding-style.md +400 -0
- package/framework/stacks/java/conventions.md +437 -0
- package/framework/stacks/java/error-handling.md +408 -0
- package/framework/stacks/java/feedback.md +180 -0
- package/framework/stacks/java/tech.md +126 -0
- package/framework/stacks/java/testing.md +485 -0
- package/framework/stacks/javascript/async-patterns.md +216 -0
- package/framework/stacks/javascript/code-quality.md +182 -0
- package/framework/stacks/javascript/coding-style.md +293 -0
- package/framework/stacks/javascript/conventions.md +268 -0
- package/framework/stacks/javascript/error-handling.md +216 -0
- package/framework/stacks/javascript/feedback.md +80 -0
- package/framework/stacks/javascript/tech.md +114 -0
- package/framework/stacks/javascript/testing.md +209 -0
- package/framework/stacks/loco/code-quality.md +156 -0
- package/framework/stacks/loco/coding-style.md +247 -0
- package/framework/stacks/loco/error-handling.md +225 -0
- package/framework/stacks/loco/feedback.md +35 -0
- package/framework/stacks/loco/loco.md +342 -0
- package/framework/stacks/loco/structure.md +193 -0
- package/framework/stacks/loco/tech.md +129 -0
- package/framework/stacks/loco/testing.md +211 -0
- package/framework/stacks/rust/code-quality.md +370 -0
- package/framework/stacks/rust/coding-style.md +475 -0
- package/framework/stacks/rust/conventions.md +430 -0
- package/framework/stacks/rust/error-handling.md +399 -0
- package/framework/stacks/rust/feedback.md +152 -0
- package/framework/stacks/rust/memory-safety.md +398 -0
- package/framework/stacks/rust/tech.md +121 -0
- package/framework/stacks/rust/testing.md +528 -0
- package/package.json +14 -2
package/README.md
CHANGED
|
@@ -1,436 +1,285 @@
|
|
|
1
1
|
# Red64 CLI
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
<div align="center">
|
|
4
4
|
|
|
5
|
-
|
|
5
|
+
### 30,000 hours of experience building software.<br/>The SDLC that makes AI-generated code maintainable.
|
|
6
6
|
|
|
7
|
-
|
|
8
|
-
<!-- TODO: Add screencast -->
|
|
7
|
+
**20 years of building products. 1+ year of AI-first development. Captured in a CLI.**
|
|
9
8
|
|
|
10
|
-
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
### 1. Install
|
|
15
|
-
|
|
16
|
-
```bash
|
|
17
|
-
# Clone and install
|
|
18
|
-
git clone https://github.com/your-org/red64-cli.git
|
|
19
|
-
cd red64-cli
|
|
20
|
-
npm install
|
|
21
|
-
npm link
|
|
22
|
-
```
|
|
9
|
+
TDD built in. Code smells to avoid. Documentation required. Quality gates enforced.<br/>
|
|
10
|
+
The process that turns AI code into production-ready software.
|
|
11
|
+
The result? Code that lives and evolves—not legacy the day it ships.
|
|
23
12
|
|
|
24
|
-
|
|
13
|
+
[](https://www.npmjs.com/package/red64-cli)
|
|
14
|
+
[](https://github.com/Red64llc/red64-cli/actions/workflows/ci.yml)
|
|
15
|
+
[](https://opensource.org/licenses/MIT)
|
|
16
|
+
[](https://red64.io/ventures)
|
|
25
17
|
|
|
26
|
-
|
|
27
|
-
cd /path/to/your/project
|
|
28
|
-
red64 init
|
|
29
|
-
```
|
|
18
|
+
[Quick Start](#-quick-start) · [Why Red64](#-why-red64) · [Features](#-features) · [Documentation](#-documentation)
|
|
30
19
|
|
|
31
|
-
|
|
32
|
-
|
|
33
|
-
```bash
|
|
34
|
-
red64 start "user-authentication" "Add login and registration with JWT tokens"
|
|
35
|
-
```
|
|
20
|
+
</div>
|
|
36
21
|
|
|
37
|
-
Red64
|
|
38
|
-
1. Create an isolated git worktree
|
|
39
|
-
2. Generate requirements (EARS format)
|
|
40
|
-
3. Create technical design
|
|
41
|
-
4. Break down into implementation tasks
|
|
42
|
-
5. Execute each task with commits
|
|
43
|
-
6. Complete with a ready-to-review branch
|
|
22
|
+
<img width="1283" height="378" alt="Red64 CLI screenshot (claude)" src="https://github.com/user-attachments/assets/cc309998-340b-4764-af83-902efc26b58f" />
|
|
44
23
|
|
|
45
24
|
---
|
|
46
25
|
|
|
47
|
-
##
|
|
26
|
+
## 🎯 The Problem
|
|
48
27
|
|
|
49
|
-
|
|
28
|
+
I've spent 20 years building products and writing software. 30,000 hours of experience. Then I went all-in on AI coding tools:
|
|
50
29
|
|
|
51
|
-
|
|
52
|
-
Requirements → Design → Tasks → Implementation
|
|
53
|
-
↓ ↓ ↓ ↓
|
|
54
|
-
(review) (review) (review) (commits)
|
|
55
|
-
```
|
|
56
|
-
|
|
57
|
-
### Key Features
|
|
30
|
+
**They're incredible for building a feature.** But then you start iterating—and you hit a wall:
|
|
58
31
|
|
|
59
|
-
|
|
60
|
-
|
|
61
|
-
|
|
62
|
-
|
|
63
|
-
|
|
64
|
-
|
|
65
|
-
|
|
66
|
-
| **Resumable Flows** | `start` auto-detects in-progress flows and offers resume |
|
|
67
|
-
| **API Health Checks** | Validates Claude API before starting (credits, auth, network) |
|
|
32
|
+
- ❌ Code quality goes down the drain
|
|
33
|
+
- ❌ No testing (or tests written after the fact)
|
|
34
|
+
- ❌ No documentation/specs (good luck iterating on anything)
|
|
35
|
+
- ❌ No careful design review, no code review
|
|
36
|
+
- ❌ No quality gates—code smells everywhere
|
|
37
|
+
- ❌ Large commits that can't be easily rolled back
|
|
38
|
+
- ❌ No non-regression tests, so things start breaking
|
|
68
39
|
|
|
69
|
-
|
|
40
|
+
**This is the same problem that arises in any team with no processes, no gates, no constraints.**
|
|
70
41
|
|
|
71
|
-
|
|
72
|
-
2. **Traceability**: Every code change links back to requirements and design decisions
|
|
73
|
-
3. **Quality**: Mandatory review gates prevent AI hallucinations from reaching production
|
|
74
|
-
4. **Isolation**: Git worktrees prevent work-in-progress from polluting main branch
|
|
75
|
-
5. **Efficiency**: Automate the tedious parts while keeping humans in control
|
|
42
|
+
## ✅ The Solution
|
|
76
43
|
|
|
77
|
-
|
|
44
|
+
The solution is what I've been doing for 20 years: **Software Development Life Cycle and Processes.** The stuff tech leaders and experience software professional implement in their teams. The stuff that separates "it works" from "it's maintainable."
|
|
78
45
|
|
|
79
|
-
|
|
46
|
+
**Red64 CLI captures both:**
|
|
80
47
|
|
|
81
|
-
|
|
48
|
+
1. **My 30,000 hours of experience** — code smells to avoid, patterns that scale, production wisdom
|
|
49
|
+
2. **My process for working with AI** — the SDLC that makes AI-generated code maintainable
|
|
82
50
|
|
|
83
|
-
|
|
84
|
-
-
|
|
85
|
-
-
|
|
86
|
-
-
|
|
51
|
+
**The process (HOW the software professional works):**
|
|
52
|
+
- Isolate every feature in a branch (git worktree)
|
|
53
|
+
- Write tests FIRST (TDD built in)
|
|
54
|
+
- Small atomic commits (one thing per commit)
|
|
55
|
+
- Document everything (REQUIREMENTS.md, DESIGN.md)
|
|
56
|
+
- High test coverage enforced
|
|
57
|
+
- Quality gates at every phase
|
|
87
58
|
|
|
88
|
-
|
|
59
|
+
**The expertise (WHAT the software professional builds):**
|
|
60
|
+
- Code smells to avoid (the stuff that breaks at 3 AM)
|
|
61
|
+
- Patterns and anti-patterns for Python, Next, Ruby, Rails etc...
|
|
62
|
+
- Stack-specific conventions (Next.js, Rails, FastAPI, etc.)
|
|
89
63
|
|
|
90
|
-
|
|
91
|
-
git clone https://github.com/your-org/red64-cli.git
|
|
92
|
-
cd red64-cli
|
|
93
|
-
npm install
|
|
94
|
-
npm run build
|
|
95
|
-
npm link
|
|
96
|
-
```
|
|
97
|
-
|
|
98
|
-
### Verify Installation
|
|
99
|
-
|
|
100
|
-
```bash
|
|
101
|
-
red64 --version
|
|
102
|
-
red64 help
|
|
103
|
-
```
|
|
64
|
+
**The result:** Code that lives and evolves. We've rewritten features in another language in **days** because the documentation is so complete.
|
|
104
65
|
|
|
105
66
|
---
|
|
106
67
|
|
|
107
|
-
##
|
|
108
|
-
|
|
109
|
-
### `red64 init`
|
|
110
|
-
|
|
111
|
-
Initialize Red64 in your project. Creates `.red64/` directory with steering documents.
|
|
68
|
+
## 🚀 Quick Start
|
|
112
69
|
|
|
113
70
|
```bash
|
|
114
|
-
|
|
115
|
-
|
|
116
|
-
red64 init --agent gemini # Use Gemini as coding agent
|
|
117
|
-
```
|
|
118
|
-
|
|
119
|
-
**Flags:**
|
|
120
|
-
- `-a, --agent <name>` — Coding agent: `claude`, `gemini`, `codex` (default: `claude`)
|
|
121
|
-
|
|
122
|
-

|
|
123
|
-
<!-- TODO: Add screenshot -->
|
|
124
|
-
|
|
125
|
-
### `red64 start <feature> <description>`
|
|
126
|
-
|
|
127
|
-
Start a new feature development flow, or resume an existing one.
|
|
128
|
-
|
|
129
|
-
```bash
|
|
130
|
-
red64 start "shopping-cart" "Add shopping cart with add/remove items and checkout"
|
|
131
|
-
```
|
|
132
|
-
|
|
133
|
-
**Smart Resume Detection:**
|
|
134
|
-
|
|
135
|
-
When you run `start` for a feature that's already in progress (at any phase: requirements, design, tasks, or implementation), Red64 will:
|
|
71
|
+
# Install
|
|
72
|
+
npm install -g red64-cli
|
|
136
73
|
|
|
137
|
-
|
|
138
|
-
|
|
139
|
-
|
|
140
|
-
- Cancel
|
|
141
|
-
|
|
142
|
-
2. **Offer resume or restart** — If an in-progress flow is found, prompts:
|
|
143
|
-
- Resume from current phase
|
|
144
|
-
- Start fresh (discard previous progress)
|
|
145
|
-
- Cancel
|
|
146
|
-
|
|
147
|
-
This means you can always use `red64 start <feature>` to continue working—no separate resume command needed.
|
|
148
|
-
|
|
149
|
-
**Flags:**
|
|
150
|
-
- `-m, --model <name>` — Model to use (e.g., `claude-3-5-haiku-latest` for dev, `claude-sonnet-4-20250514` for prod)
|
|
151
|
-
- `-y, --yes` — Auto-approve all phases (skip review gates)
|
|
152
|
-
- `-b, --brownfield` — Enable gap analysis for existing codebases
|
|
153
|
-
- `-g, --greenfield` — New feature mode (default)
|
|
154
|
-
- `-s, --skip-permissions` — Pass to Claude CLI
|
|
155
|
-
- `-t, --tier <name>` — Use specific Claude config tier
|
|
156
|
-
- `--sandbox` — Run in Docker isolation
|
|
157
|
-
- `--verbose` — Show detailed execution logs
|
|
158
|
-
|
|
159
|
-
<img width="1134" height="567" alt="Screenshot 2026-01-29 at 6 35 01 PM" src="https://github.com/user-attachments/assets/91ea478f-53de-4964-a0b2-b0f05d2bfb2b" />
|
|
160
|
-
|
|
161
|
-
### `red64 status [feature]`
|
|
74
|
+
# Initialize in your project
|
|
75
|
+
cd /path/to/your/project
|
|
76
|
+
red64 init --stack nextjs
|
|
162
77
|
|
|
163
|
-
|
|
78
|
+
# Start a feature (interactive mode)
|
|
79
|
+
red64 start "user-auth" "Add login and registration with JWT"
|
|
164
80
|
|
|
165
|
-
|
|
166
|
-
red64
|
|
167
|
-
red64 status # Show all flows
|
|
81
|
+
# Or YOLO mode — no babysitting required
|
|
82
|
+
red64 start "shopping-cart" "Full cart with checkout" --sandbox -y
|
|
168
83
|
```
|
|
169
84
|
|
|
170
|
-
|
|
171
|
-
<!-- TODO: Add screenshot -->
|
|
172
|
-
|
|
173
|
-
### `red64 list`
|
|
85
|
+
That's it. Red64 generates requirements → design → tests → implementation → documentation.
|
|
174
86
|
|
|
175
|
-
|
|
87
|
+
Each phase has review checkpoints. Each task = one clean commit. Tests first. Docs included.
|
|
176
88
|
|
|
177
|
-
|
|
178
|
-
red64 list
|
|
179
|
-
```
|
|
89
|
+
---
|
|
180
90
|
|
|
181
|
-
|
|
91
|
+
## 🔥 YOLO Mode (No Babysitting)
|
|
182
92
|
|
|
183
|
-
|
|
93
|
+
Tired of approving every line?
|
|
184
94
|
|
|
185
95
|
```bash
|
|
186
|
-
red64
|
|
96
|
+
red64 start "feature-name" "description" --sandbox -y
|
|
187
97
|
```
|
|
188
98
|
|
|
189
|
-
|
|
190
|
-
|
|
191
|
-
## Workflow Phases
|
|
192
|
-
|
|
193
|
-
### 1. Initialization
|
|
194
|
-
|
|
195
|
-
Creates spec directory at `.red64/specs/<feature>/` with `spec.json`.
|
|
196
|
-
|
|
197
|
-
### 2. Requirements Generation
|
|
198
|
-
|
|
199
|
-
Generates `requirements.md` using EARS (Easy Approach to Requirements Syntax) format:
|
|
200
|
-
- Ubiquitous requirements
|
|
201
|
-
- Event-driven requirements
|
|
202
|
-
- State-driven requirements
|
|
203
|
-
- Optional features
|
|
204
|
-
- Unwanted behaviors
|
|
205
|
-
|
|
206
|
-
### 3. Gap Analysis (Brownfield only)
|
|
207
|
-
|
|
208
|
-
For existing codebases, analyzes what already exists and what needs to be built.
|
|
209
|
-
|
|
210
|
-
### 4. Design Generation
|
|
211
|
-
|
|
212
|
-
Creates `design.md` with:
|
|
213
|
-
- Architecture decisions
|
|
214
|
-
- Component design
|
|
215
|
-
- Data models
|
|
216
|
-
- API contracts
|
|
217
|
-
- File structure
|
|
218
|
-
|
|
219
|
-
### 5. Task Generation
|
|
220
|
-
|
|
221
|
-
Breaks down the design into atomic `tasks.md`:
|
|
222
|
-
- Each task is independently implementable
|
|
223
|
-
- Tasks are ordered by dependency
|
|
224
|
-
- Each task produces a commit
|
|
99
|
+
- `--sandbox` = Docker isolation (AI can't break your system, pulls image from `ghcr.io/red64llc/red64-sandbox`)
|
|
100
|
+
- `-y` = Auto-approve all phases (total autonomy)
|
|
225
101
|
|
|
226
|
-
|
|
102
|
+
**Start a feature. Go to lunch. Come back to a completed branch—with tests, docs, and clean commits.**
|
|
227
103
|
|
|
228
|
-
|
|
229
|
-
|
|
230
|
-
- Commits after each task
|
|
231
|
-
- Checkpoints every 3 tasks (optional pause)
|
|
104
|
+
With other tools, YOLO mode means "write code fast with no oversight."<br/>
|
|
105
|
+
With Red64, autonomous mode means "follow the SDLC with no babysitting."
|
|
232
106
|
|
|
233
|
-
|
|
107
|
+
The AI still:
|
|
108
|
+
1. Writes tests FIRST (TDD enforced)
|
|
109
|
+
2. Documents everything (REQUIREMENTS.md, DESIGN.md)
|
|
110
|
+
3. Makes atomic commits (easy to review, easy to rollback)
|
|
111
|
+
4. Passes quality gates (no code smells ship)
|
|
234
112
|
|
|
235
|
-
|
|
236
|
-
- All tasks implemented
|
|
237
|
-
- Clean commit history
|
|
238
|
-
- Feature branch ready for PR
|
|
113
|
+
**Review the PR when it's done. Like a senior engineer delegating to a junior who's been properly onboarded.**
|
|
239
114
|
|
|
240
115
|
---
|
|
241
116
|
|
|
242
|
-
##
|
|
117
|
+
## 🏆 Battle-Tested
|
|
243
118
|
|
|
244
|
-
|
|
245
|
-
|---------|-------|----------------|-------|-------------|
|
|
246
|
-
| **Spec-driven workflow** | Yes | No | No | No |
|
|
247
|
-
| **Phase gates** | Yes | No | No | No |
|
|
248
|
-
| **Git worktree isolation** | Yes | No | No | No |
|
|
249
|
-
| **Atomic commits per task** | Yes | No | Yes | No |
|
|
250
|
-
| **Requirements generation** | Yes | No | No | No |
|
|
251
|
-
| **Design documents** | Yes | No | No | No |
|
|
252
|
-
| **Resumable flows** | Yes | No | Partial | No |
|
|
253
|
-
| **Docker sandboxing** | Yes | No | No | Yes |
|
|
254
|
-
| **IDE integration** | No | Yes | Partial | Yes |
|
|
255
|
-
| **Real-time editing** | No | Yes | Yes | Yes |
|
|
119
|
+
We built **6 production products** with Red64 at [red64.io/ventures](https://red64.io/ventures):
|
|
256
120
|
|
|
257
|
-
|
|
121
|
+
| Company | Industry | Status |
|
|
122
|
+
|---------|----------|--------|
|
|
123
|
+
| [Saife](https://red64.io/ventures) | InsurTech | Production |
|
|
124
|
+
| [EngineValue](https://red64.io/ventures) | Engineering Scorecards | Production |
|
|
125
|
+
| [MediaPulse](https://red64.io/ventures) | Digital Presence | Production |
|
|
126
|
+
| [TheScriptMarketplace](https://red64.io/ventures) | Entertainment | Production |
|
|
127
|
+
| [QueryVault](https://red64.io/ventures) | Data Platform | Production |
|
|
128
|
+
| [Kafi](Internal product) | Virtual Executive Assistant | Production |
|
|
258
129
|
|
|
259
|
-
|
|
260
|
-
- Building complete features (not quick fixes)
|
|
261
|
-
- You need traceable requirements and design
|
|
262
|
-
- Working on complex, multi-file changes
|
|
263
|
-
- You want clean git history with atomic commits
|
|
264
|
-
- You need to pause and resume work across sessions
|
|
265
|
-
|
|
266
|
-
**Use other tools when:**
|
|
267
|
-
- Making quick, single-file edits
|
|
268
|
-
- Exploring or prototyping ideas
|
|
269
|
-
- You need real-time IDE integration
|
|
270
|
-
- Working on bug fixes without spec requirements
|
|
130
|
+
Same tool. Same encoded experience. Now open source.
|
|
271
131
|
|
|
272
132
|
---
|
|
273
133
|
|
|
274
|
-
##
|
|
134
|
+
## 💡 Why Red64?
|
|
275
135
|
|
|
276
|
-
###
|
|
136
|
+
### Two Decades of Experience, Encoded
|
|
277
137
|
|
|
278
|
-
|
|
279
|
-
2. **Sequential execution** — Tasks run one at a time, no parallelization
|
|
280
|
-
3. **No incremental changes** — Regenerating a phase replaces previous output entirely
|
|
281
|
-
4. **English-centric** — Prompts and templates are English-only (configurable per spec)
|
|
138
|
+
I've spent 20 years building products—30,000 hours of learning what works and what breaks. Then I spent a year going all-in on AI coding tools.
|
|
282
139
|
|
|
283
|
-
|
|
140
|
+
**The pattern is always the same:**
|
|
284
141
|
|
|
285
|
-
|
|
286
|
-
|
|
287
|
-
|
|
142
|
+
1. **Week 1:** "This is amazing! I shipped a feature in a day!"
|
|
143
|
+
2. **Week 4:** "Why is everything breaking? Why is the code so messy?"
|
|
144
|
+
3. **Week 8:** "I'm afraid to touch anything. Time to rewrite."
|
|
288
145
|
|
|
289
|
-
|
|
146
|
+
**The missing ingredient?** SDLC. The stuff that takes 20 years to learn. The stuff I've been teaching engineers my entire career.
|
|
290
147
|
|
|
291
|
-
|
|
292
|
-
- [ ] Parallel task execution
|
|
293
|
-
- [ ] Incremental phase editing
|
|
294
|
-
- [ ] Web dashboard for flow monitoring
|
|
148
|
+
Red64 gives you both:
|
|
295
149
|
|
|
296
|
-
|
|
150
|
+
| What Goes Wrong Without SDLC | Red64 Solution |
|
|
151
|
+
|------------------------------|----------------|
|
|
152
|
+
| No tests → things break when you iterate | TDD built in (tests FIRST) |
|
|
153
|
+
| No docs → can't remember why anything works | REQUIREMENTS.md + DESIGN.md per feature |
|
|
154
|
+
| Huge commits → can't rollback, can't review | Atomic commits (one task = one commit) |
|
|
155
|
+
| No quality gates → code smells everywhere | Guardrails from 30K hours of experience |
|
|
156
|
+
| Babysitting every line → slow, exhausting | Autonomous mode with SDLC guardrails |
|
|
297
157
|
|
|
298
|
-
|
|
158
|
+
### What You Get Per Feature
|
|
299
159
|
|
|
300
160
|
```
|
|
301
|
-
|
|
302
|
-
├──
|
|
303
|
-
|
|
304
|
-
|
|
305
|
-
|
|
306
|
-
│
|
|
307
|
-
│
|
|
308
|
-
|
|
309
|
-
|
|
310
|
-
│ ├── state.json # Current flow state
|
|
311
|
-
│ └── flow.log # Execution log
|
|
312
|
-
└── steering/ # Project-wide guidance
|
|
313
|
-
├── product.md # Product context
|
|
314
|
-
├── tech.md # Technical standards
|
|
315
|
-
└── structure.md # Codebase structure
|
|
161
|
+
feature-branch/
|
|
162
|
+
├── REQUIREMENTS.md # What we're building and why
|
|
163
|
+
├── DESIGN.md # How it works, architecture decisions
|
|
164
|
+
├── TASKS.md # Atomic breakdown with acceptance criteria
|
|
165
|
+
├── src/
|
|
166
|
+
│ ├── feature.ts # Implementation
|
|
167
|
+
│ └── feature.test.ts # Tests (written first)
|
|
168
|
+
└── docs/
|
|
169
|
+
└── feature.md # User-facing documentation
|
|
316
170
|
```
|
|
317
171
|
|
|
318
|
-
|
|
319
|
-
|
|
320
|
-
Red64 creates git worktrees in a **sibling directory** following the [GitLens/VSCode convention](https://gist.github.com/ChristopherA/4643b2f5e024578606b9cd5d2e6815cc). This keeps the main repository clean and avoids nested git structures.
|
|
321
|
-
|
|
322
|
-
```
|
|
323
|
-
~/projects/
|
|
324
|
-
├── my-project/ # Main repository (unchanged)
|
|
325
|
-
│ ├── .git/
|
|
326
|
-
│ ├── .red64/
|
|
327
|
-
│ ├── src/
|
|
328
|
-
│ └── package.json
|
|
329
|
-
└── my-project.worktrees/ # Worktrees directory (sibling)
|
|
330
|
-
├── user-authentication/ # feature/user-authentication branch
|
|
331
|
-
│ ├── .red64/
|
|
332
|
-
│ ├── src/
|
|
333
|
-
│ └── package.json
|
|
334
|
-
└── shopping-cart/ # feature/shopping-cart branch
|
|
335
|
-
├── .red64/
|
|
336
|
-
├── src/
|
|
337
|
-
└── package.json
|
|
338
|
-
```
|
|
339
|
-
|
|
340
|
-
**Benefits of sibling worktrees:**
|
|
341
|
-
- Main repository stays clean (no `.gitignore` needed for worktrees)
|
|
342
|
-
- No nested `.git` references inside the working tree
|
|
343
|
-
- Better IDE and tool compatibility
|
|
344
|
-
- Clear separation between main work and feature isolation
|
|
345
|
-
- Easy to see all active features at a glance
|
|
172
|
+
Every decision traceable. Every line has a reason. **Code that survives iteration.**
|
|
346
173
|
|
|
347
174
|
---
|
|
348
175
|
|
|
349
|
-
##
|
|
176
|
+
## 📊 Comparison
|
|
177
|
+
|
|
178
|
+
| Feature | Red64 | Cursor | Copilot | Claude Code | Gemini CLI | Aider |
|
|
179
|
+
|---------|:-----:|:------:|:-------:|:-----------:|:----------:|:-----:|
|
|
180
|
+
| **30K hours expertise encoded** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
181
|
+
| **SDLC/Process enforced** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
182
|
+
| **Autonomous mode** | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
|
|
183
|
+
| **Sandboxed execution** | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ |
|
|
184
|
+
| **MCP support** | ✅ | ⚠️ | ✅ | ✅ | ✅ | ❌ |
|
|
185
|
+
| **TDD enforced (tests first)** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
186
|
+
| **High coverage enforced** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
187
|
+
| **Auto-generates docs** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
188
|
+
| **Git worktree isolation** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
189
|
+
| **Atomic commits enforced** | ✅ | ❌ | ❌ | ❌ | ❌ | ✅ |
|
|
190
|
+
| **Phase gates with review** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
191
|
+
| **Code smell guardrails** | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
192
|
+
| **Resumable multi-step flows** | ✅ | ❌ | ❌ | ✅ | ❌ | ⚠️ |
|
|
193
|
+
| **Multi-model support** | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ |
|
|
194
|
+
| **Battle-tested (production)** | ✅ 6 cos | N/A | N/A | N/A | N/A | N/A |
|
|
195
|
+
|
|
196
|
+
**Key:** ✅ = Built-in & enforced | ⚠️ = Partial/Optional | ❌ = Not available
|
|
197
|
+
|
|
198
|
+
> **The difference:** Other tools have autonomous modes. Red64 has autonomous mode **plus** the encoded expertise and enforced process that produces production-quality code.
|
|
199
|
+
|
|
200
|
+
### When to Use Red64
|
|
201
|
+
|
|
202
|
+
✅ **Use Red64 when:**
|
|
203
|
+
- Building complete features (not quick fixes)
|
|
204
|
+
- You want code with tests, docs, and clean history
|
|
205
|
+
- You need to walk away and let AI work autonomously
|
|
206
|
+
- You're tired of babysitting every line
|
|
207
|
+
- You want code that's safe to refactor
|
|
350
208
|
|
|
351
|
-
|
|
209
|
+
❌ **Use other tools when:**
|
|
210
|
+
- Making quick, single-file edits
|
|
211
|
+
- You want real-time IDE autocomplete
|
|
212
|
+
- Exploring or prototyping ideas
|
|
352
213
|
|
|
353
|
-
|
|
214
|
+
---
|
|
354
215
|
|
|
355
|
-
|
|
356
|
-
- **tech.md** — Tech stack, coding standards, patterns to use/avoid
|
|
357
|
-
- **structure.md** — Codebase organization, file naming conventions
|
|
216
|
+
## ⚡ Features
|
|
358
217
|
|
|
359
|
-
###
|
|
218
|
+
### Multi-Agent Support
|
|
360
219
|
|
|
361
|
-
|
|
220
|
+
Use your preferred AI:
|
|
362
221
|
|
|
363
222
|
```bash
|
|
364
|
-
red64 init --agent claude # Default
|
|
223
|
+
red64 init --agent claude # Default
|
|
365
224
|
red64 init --agent gemini # Google Gemini
|
|
366
225
|
red64 init --agent codex # OpenAI Codex
|
|
367
226
|
```
|
|
368
227
|
|
|
369
|
-
|
|
370
|
-
|
|
371
|
-
### Model Selection
|
|
228
|
+
### Smart Resume
|
|
372
229
|
|
|
373
|
-
|
|
230
|
+
Interrupted? Just run `start` again:
|
|
374
231
|
|
|
375
232
|
```bash
|
|
376
|
-
|
|
377
|
-
|
|
378
|
-
red64 start "feature" "desc" --model gemini-2.0-flash
|
|
379
|
-
red64 start "feature" "desc" --model gpt-4o-mini
|
|
380
|
-
|
|
381
|
-
# Production (best quality models)
|
|
382
|
-
red64 start "feature" "desc" --model claude-sonnet-4-20250514
|
|
383
|
-
red64 start "feature" "desc" --model gemini-2.5-pro
|
|
384
|
-
red64 start "feature" "desc" --model o1
|
|
233
|
+
red64 start "shopping-cart" "..."
|
|
234
|
+
# Detects in-progress flow, offers to resume
|
|
385
235
|
```
|
|
386
236
|
|
|
387
|
-
|
|
388
|
-
|-------|-------------|-------------|
|
|
389
|
-
| Claude | `claude-3-5-haiku-latest` | `claude-sonnet-4-20250514` |
|
|
390
|
-
| Gemini | `gemini-2.0-flash` | `gemini-2.5-pro` |
|
|
391
|
-
| Codex | `gpt-4o-mini` | `o1` |
|
|
392
|
-
|
|
393
|
-
### Claude Tiers
|
|
237
|
+
### Steering Documents
|
|
394
238
|
|
|
395
|
-
|
|
239
|
+
Customize AI behavior in `.red64/steering/`:
|
|
396
240
|
|
|
397
|
-
|
|
398
|
-
|
|
399
|
-
|
|
400
|
-
```
|
|
241
|
+
- **product.md** — Product vision, user personas
|
|
242
|
+
- **tech.md** — Stack standards, code smells to avoid
|
|
243
|
+
- **structure.md** — Codebase organization
|
|
401
244
|
|
|
402
245
|
---
|
|
403
246
|
|
|
404
|
-
##
|
|
405
|
-
|
|
406
|
-
### Run in development mode
|
|
407
|
-
|
|
408
|
-
```bash
|
|
409
|
-
npm run dev -- start my-feature "Feature description"
|
|
410
|
-
```
|
|
247
|
+
## 📖 Documentation
|
|
411
248
|
|
|
412
|
-
|
|
249
|
+
- [Full Documentation](./docs/README.md)
|
|
250
|
+
- [Steering Document Guide](./docs/steering.md)
|
|
251
|
+
- [Configuration Reference](./docs/configuration.md)
|
|
252
|
+
- [Troubleshooting](./docs/troubleshooting.md)
|
|
413
253
|
|
|
414
|
-
|
|
415
|
-
npm test
|
|
416
|
-
npm run test:ui
|
|
417
|
-
```
|
|
254
|
+
---
|
|
418
255
|
|
|
419
|
-
|
|
256
|
+
## 🛠 Commands
|
|
420
257
|
|
|
421
258
|
```bash
|
|
422
|
-
|
|
259
|
+
red64 init --agent gemini # Initialize Red64 in your project
|
|
260
|
+
red64 start <feature> <desc> # Start a new feature
|
|
261
|
+
red64 start ... --sandbox -y # YOLO mode (autonomous)
|
|
262
|
+
red64 status [feature] # Check flow status
|
|
263
|
+
red64 list # List all active flows
|
|
264
|
+
red64 abort <feature> # Abort and clean up
|
|
423
265
|
```
|
|
424
266
|
|
|
425
|
-
###
|
|
267
|
+
### Flags
|
|
426
268
|
|
|
427
|
-
|
|
428
|
-
|
|
429
|
-
|
|
269
|
+
| Flag | Description |
|
|
270
|
+
|------|-------------|
|
|
271
|
+
| `-y, --yes` | Auto-approve all phases (YOLO mode) |
|
|
272
|
+
| `--sandbox` | Run in Docker isolation (uses GHCR image by default) |
|
|
273
|
+
| `--local-image` | Build and use local sandbox image instead of GHCR (init only) |
|
|
274
|
+
| `-m, --model` | Override AI model |
|
|
275
|
+
| `-a, --agent` | Set coding agent (claude/gemini/codex) |
|
|
276
|
+
| `--verbose` | Show detailed logs |
|
|
430
277
|
|
|
431
278
|
---
|
|
432
279
|
|
|
433
|
-
## Contributing
|
|
280
|
+
## 🤝 Contributing
|
|
281
|
+
|
|
282
|
+
We'd love your help encoding more production wisdom:
|
|
434
283
|
|
|
435
284
|
1. Fork the repository
|
|
436
285
|
2. Create a feature branch
|
|
@@ -438,16 +287,23 @@ npm run type-check
|
|
|
438
287
|
4. Run tests: `npm test`
|
|
439
288
|
5. Submit a pull request
|
|
440
289
|
|
|
290
|
+
**What we're looking for:**
|
|
291
|
+
- More code smells to catch
|
|
292
|
+
- Stack-specific best practices
|
|
293
|
+
- Bug fixes and improvements
|
|
294
|
+
|
|
441
295
|
---
|
|
442
296
|
|
|
443
|
-
## License
|
|
297
|
+
## 📜 License
|
|
444
298
|
|
|
445
|
-
MIT
|
|
299
|
+
MIT — Built by [Yacin Bahi](mailto:yacin@red64.io) at [Red64.io](https://red64.io)
|
|
446
300
|
|
|
447
301
|
---
|
|
448
302
|
|
|
449
|
-
|
|
303
|
+
<div align="center">
|
|
304
|
+
|
|
305
|
+
### The code isn't the asset.<br/>The documentation + tests + history is the asset.<br/>The code is just the current implementation.
|
|
306
|
+
|
|
307
|
+
**[⭐ Star this repo](https://github.com/Red64llc/red64-cli)** if you believe AI should write code like a senior engineer.
|
|
450
308
|
|
|
451
|
-
|
|
452
|
-
- Inspired by spec-driven development and EARS requirements methodology
|
|
453
|
-
- Uses [Ink](https://github.com/vadimdemedes/ink) for terminal UI
|
|
309
|
+
</div>
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"parseArgs.d.ts","sourceRoot":"","sources":["../../src/cli/parseArgs.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,KAAK,EAAE,SAAS,EAAqC,MAAM,mBAAmB,CAAC;AAatF;;;GAGG;AACH,wBAAgB,SAAS,CAAC,IAAI,EAAE,SAAS,MAAM,EAAE,GAAG,SAAS,
|
|
1
|
+
{"version":3,"file":"parseArgs.d.ts","sourceRoot":"","sources":["../../src/cli/parseArgs.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,KAAK,EAAE,SAAS,EAAqC,MAAM,mBAAmB,CAAC;AAatF;;;GAGG;AACH,wBAAgB,SAAS,CAAC,IAAI,EAAE,SAAS,MAAM,EAAE,GAAG,SAAS,CA0G5D;AAED;;;GAGG;AACH,eAAO,MAAM,SAAS,QAyBd,CAAC"}
|