gaia-framework 1.27.0 → 1.27.2
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 +192 -497
- package/gaia-install.sh +1 -1
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -1,10 +1,73 @@
|
|
|
1
1
|
# GAIA — Generative Agile Intelligence Architecture
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
[]()
|
|
4
|
+
[]()
|
|
5
|
+
[]()
|
|
6
|
+
[]()
|
|
4
7
|
|
|
5
|
-
AI agent framework for [Claude Code](https://docs.anthropic.com/en/docs/claude-code) that orchestrates software product development through 25 specialized agents
|
|
8
|
+
AI agent framework for [Claude Code](https://docs.anthropic.com/en/docs/claude-code) that orchestrates software product development through **25 specialized agents**, **64 workflows**, and **8 shared skills** — from initial research all the way to deployment.
|
|
6
9
|
|
|
7
|
-
GAIA
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
### Why GAIA?
|
|
13
|
+
|
|
14
|
+
Using Claude Code alone, you prompt an AI assistant. With GAIA, you manage a **team of AI specialists** that follow a proven product lifecycle:
|
|
15
|
+
|
|
16
|
+
- **Structured lifecycle** — 5 phases from analysis to deployment, with quality gates that enforce standards at every transition
|
|
17
|
+
- **25 specialized agents** — each with a persona, domain expertise, and persistent memory that improves over time
|
|
18
|
+
- **Built-in quality gates** — 17 enforced gates that HALT workflows when standards aren't met (not advisory — hard stops)
|
|
19
|
+
- **6-gate review process** — every story passes code review, QA, security, test automation, test review, and performance review before completion
|
|
20
|
+
- **Checkpoint/resume** — long-running workflows survive context loss with sha256-verified checkpoints
|
|
21
|
+
- **Brownfield support** — onboard existing codebases with automated discovery, documentation generation, and gap analysis
|
|
22
|
+
|
|
23
|
+
---
|
|
24
|
+
|
|
25
|
+
## Prerequisites
|
|
26
|
+
|
|
27
|
+
- [Claude Code](https://docs.anthropic.com/en/docs/claude-code) (requires Anthropic account)
|
|
28
|
+
- Node.js 18+
|
|
29
|
+
- Git
|
|
30
|
+
|
|
31
|
+
---
|
|
32
|
+
|
|
33
|
+
## Quick Start
|
|
34
|
+
|
|
35
|
+
```bash
|
|
36
|
+
# Install GAIA into your project
|
|
37
|
+
npx gaia-framework init ~/my-project
|
|
38
|
+
|
|
39
|
+
# Open in Claude Code
|
|
40
|
+
cd ~/my-project && claude
|
|
41
|
+
|
|
42
|
+
# Generate pre-resolved configs (one-time setup)
|
|
43
|
+
/gaia-build-configs
|
|
44
|
+
```
|
|
45
|
+
|
|
46
|
+
Then try your first workflow:
|
|
47
|
+
|
|
48
|
+
```bash
|
|
49
|
+
# Describe a small feature and get a tech spec in 30 seconds
|
|
50
|
+
/gaia-quick-spec
|
|
51
|
+
|
|
52
|
+
# Implement it immediately
|
|
53
|
+
/gaia-quick-dev
|
|
54
|
+
```
|
|
55
|
+
|
|
56
|
+
Or launch the orchestrator to explore all capabilities:
|
|
57
|
+
|
|
58
|
+
```bash
|
|
59
|
+
/gaia
|
|
60
|
+
```
|
|
61
|
+
|
|
62
|
+
### 5 essential commands
|
|
63
|
+
|
|
64
|
+
| Command | What it does |
|
|
65
|
+
|---------|-------------|
|
|
66
|
+
| `/gaia` | Launch the orchestrator — shows categories and routes you |
|
|
67
|
+
| `/gaia-dev-story` | Implement a user story end-to-end |
|
|
68
|
+
| `/gaia-quick-spec` | Create a rapid tech spec for small changes |
|
|
69
|
+
| `/gaia-quick-dev` | Implement a quick spec |
|
|
70
|
+
| `/gaia-help` | Context-sensitive help for wherever you are |
|
|
8
71
|
|
|
9
72
|
---
|
|
10
73
|
|
|
@@ -16,179 +79,87 @@ GAIA gives you a team of AI agents with distinct personas, structured workflows
|
|
|
16
79
|
npx gaia-framework init .
|
|
17
80
|
```
|
|
18
81
|
|
|
19
|
-
This clones the latest GAIA release from GitHub and installs it into the current directory.
|
|
20
|
-
|
|
21
82
|
```bash
|
|
22
|
-
# Install into a
|
|
83
|
+
# Install into a specific directory
|
|
23
84
|
npx gaia-framework init ~/my-new-project
|
|
24
85
|
|
|
25
|
-
# Skip interactive prompts
|
|
86
|
+
# Skip interactive prompts
|
|
26
87
|
npx gaia-framework init --yes ~/my-project
|
|
27
88
|
```
|
|
28
89
|
|
|
29
|
-
|
|
30
|
-
|
|
31
|
-
### Using the shell script directly
|
|
90
|
+
### Using the shell script
|
|
32
91
|
|
|
33
92
|
```bash
|
|
34
|
-
# Clone the repo
|
|
35
93
|
git clone https://github.com/J-louage/Gaia-framework.git
|
|
36
|
-
|
|
37
|
-
# Install into your project
|
|
38
94
|
bash Gaia-framework/gaia-install.sh init ~/my-project
|
|
39
|
-
|
|
40
|
-
# Or use a local copy as source
|
|
41
|
-
bash gaia-install.sh init --source ~/Gaia-framework ~/my-project
|
|
42
95
|
```
|
|
43
96
|
|
|
44
|
-
The script resolves sources in this order: `--source` flag, `$GAIA_SOURCE` env var, script's own directory, GitHub clone.
|
|
45
|
-
|
|
46
97
|
### What the installer does
|
|
47
98
|
|
|
48
99
|
1. Copies the `_gaia/` framework into your project
|
|
49
100
|
2. Creates `docs/` artifact directories (planning, implementation, test, creative)
|
|
50
|
-
3. Creates memory sidecar directories
|
|
51
|
-
4.
|
|
52
|
-
5.
|
|
53
|
-
6.
|
|
54
|
-
7.
|
|
55
|
-
8. Appends GAIA entries to `.gitignore`
|
|
101
|
+
3. Creates memory sidecar directories
|
|
102
|
+
4. Prompts for project name and user name → writes to `global.yaml`
|
|
103
|
+
5. Copies `CLAUDE.md` to your project root
|
|
104
|
+
6. Installs 104 slash commands to `.claude/commands/`
|
|
105
|
+
7. Appends GAIA entries to `.gitignore`
|
|
56
106
|
|
|
57
|
-
### Updating
|
|
107
|
+
### Updating
|
|
58
108
|
|
|
59
109
|
```bash
|
|
60
|
-
# Via npx
|
|
61
110
|
npx gaia-framework update .
|
|
62
|
-
|
|
63
|
-
# Via shell script
|
|
64
|
-
bash gaia-install.sh update ~/my-project
|
|
65
111
|
```
|
|
66
112
|
|
|
67
|
-
Update refreshes all framework files
|
|
68
|
-
|
|
69
|
-
Changed files are backed up to `_gaia/_backups/{timestamp}/` before overwriting.
|
|
113
|
+
Update refreshes all framework files while preserving your configuration, agent memory, and `CLAUDE.md`. Changed files are backed up to `_gaia/_backups/{timestamp}/`.
|
|
70
114
|
|
|
71
|
-
### Validating
|
|
115
|
+
### Validating
|
|
72
116
|
|
|
73
117
|
```bash
|
|
74
|
-
|
|
75
|
-
npx gaia-framework
|
|
76
|
-
|
|
77
|
-
# Show version, module list, command count, sidecar status
|
|
78
|
-
npx gaia-framework status .
|
|
79
|
-
```
|
|
80
|
-
|
|
81
|
-
### Installer options
|
|
82
|
-
|
|
83
|
-
```
|
|
84
|
-
gaia-install.sh <command> [options] [target]
|
|
85
|
-
|
|
86
|
-
Commands:
|
|
87
|
-
init Install GAIA into a project
|
|
88
|
-
update Update framework files (preserves config and memory)
|
|
89
|
-
validate Check installation integrity
|
|
90
|
-
status Show installation info
|
|
91
|
-
|
|
92
|
-
Options:
|
|
93
|
-
--source <path> Local GAIA source directory
|
|
94
|
-
--yes Skip confirmation prompts
|
|
95
|
-
--dry-run Show what would be done without making changes
|
|
96
|
-
--verbose Show detailed progress
|
|
97
|
-
--help Show help
|
|
118
|
+
npx gaia-framework validate . # 32 integrity checks
|
|
119
|
+
npx gaia-framework status . # version, module list, command count
|
|
98
120
|
```
|
|
99
121
|
|
|
100
122
|
---
|
|
101
123
|
|
|
102
|
-
##
|
|
124
|
+
## How It Works
|
|
103
125
|
|
|
104
|
-
|
|
126
|
+
### 5-phase product lifecycle
|
|
105
127
|
|
|
106
|
-
```
|
|
107
|
-
|
|
108
|
-
|
|
128
|
+
```
|
|
129
|
+
Phase 1: Analysis → Phase 2: Planning → Phase 3: Solutioning → Phase 4: Implementation → Phase 5: Deployment
|
|
130
|
+
(Elena) (Derek/Christy) (Theo/Zara/Soren) (Nate/Dev Agents) (Soren)
|
|
109
131
|
```
|
|
110
132
|
|
|
111
|
-
|
|
133
|
+
Every phase has **quality gates** — enforced checks that halt the workflow if prerequisites aren't met. You can't create an architecture without a reviewed PRD. You can't start a sprint without a readiness check. You can't deploy without passing all reviews.
|
|
112
134
|
|
|
113
|
-
###
|
|
135
|
+
### Entry points
|
|
114
136
|
|
|
115
|
-
| Command |
|
|
116
|
-
|
|
117
|
-
|
|
|
118
|
-
| `/gaia-
|
|
119
|
-
| `/gaia-quick-spec` |
|
|
120
|
-
| `/gaia-
|
|
121
|
-
| `/gaia-help` | Context-sensitive help for wherever you are |
|
|
137
|
+
| Path | Command | When to use |
|
|
138
|
+
|------|---------|-------------|
|
|
139
|
+
| New project | Start at Phase 1 | Greenfield — building from scratch |
|
|
140
|
+
| Existing project | `/gaia-brownfield` | Brownfield — onboarding existing codebase |
|
|
141
|
+
| Small change | `/gaia-quick-spec` | Under 5 files, under 1 day |
|
|
142
|
+
| Resume | `/gaia-resume` | Continue after context loss |
|
|
122
143
|
|
|
123
|
-
###
|
|
144
|
+
### Model assignment
|
|
124
145
|
|
|
125
|
-
|
|
146
|
+
Each command declares which Claude model to use. **Opus** handles deep reasoning, architectural decisions, and complex analysis (27 commands). **Sonnet** handles structured generation, template-following, and status reporting (77 commands). Claude Code auto-selects the model per workflow.
|
|
126
147
|
|
|
127
|
-
|
|
148
|
+
### Execution modes
|
|
128
149
|
|
|
129
|
-
|
|
130
|
-
|
|
131
|
-
```
|
|
132
|
-
_gaia/
|
|
133
|
-
├── _config/ # Global config, manifests, agent registry
|
|
134
|
-
│ ├── global.yaml # Project settings — single source of truth
|
|
135
|
-
│ └── manifest.yaml # Module registry
|
|
136
|
-
├── _memory/ # Persistent agent memory + checkpoints
|
|
137
|
-
│ ├── checkpoints/ # Workflow progress snapshots
|
|
138
|
-
│ └── *-sidecar/ # Per-agent persistent memory (9 sidecars)
|
|
139
|
-
├── core/ # Execution engine, protocols, shared tasks
|
|
140
|
-
│ ├── engine/ # workflow.xml, task-runner.xml, error-recovery.xml
|
|
141
|
-
│ ├── tasks/ # 16 standalone review/utility tasks
|
|
142
|
-
│ └── workflows/ # Brainstorming, party mode, advanced elicitation
|
|
143
|
-
├── lifecycle/ # Product lifecycle (5 phases)
|
|
144
|
-
│ ├── agents/ # 11 lifecycle agents
|
|
145
|
-
│ ├── workflows/ # 36 workflows across 5 phases + anytime + quick-flow
|
|
146
|
-
│ ├── templates/ # 16 document templates (PRD, architecture, API docs, brownfield, etc.)
|
|
147
|
-
│ └── teams/ # Pre-built team compositions
|
|
148
|
-
├── dev/ # Developer tooling
|
|
149
|
-
│ ├── agents/ # 6 stack-specific developers + base
|
|
150
|
-
│ ├── skills/ # 8 shared skills with sectioned loading
|
|
151
|
-
│ └── knowledge/ # Stack-specific patterns (Angular, React, Flutter, etc.)
|
|
152
|
-
├── creative/ # Creative intelligence
|
|
153
|
-
│ ├── agents/ # 6 creative agents
|
|
154
|
-
│ ├── workflows/ # 7 creative workflows
|
|
155
|
-
│ └── data/ # Methods, frameworks, story types
|
|
156
|
-
└── testing/ # Test architecture
|
|
157
|
-
├── agents/ # Test Architect (Sable)
|
|
158
|
-
├── workflows/ # 12 testing workflows
|
|
159
|
-
└── knowledge/ # 20+ testing knowledge fragments
|
|
160
|
-
```
|
|
161
|
-
|
|
162
|
-
### At a glance
|
|
163
|
-
|
|
164
|
-
| Component | Count |
|
|
165
|
-
|-----------|-------|
|
|
166
|
-
| Modules | 5 (core, lifecycle, dev, creative, testing) |
|
|
167
|
-
| Agents | 25 with distinct personas |
|
|
168
|
-
| Workflows | 59 covering the full product lifecycle |
|
|
169
|
-
| Standalone tasks | 16 (reviews, audits, utilities) |
|
|
170
|
-
| Shared skills | 8 with 47 loadable sections |
|
|
171
|
-
| Slash commands | 100 |
|
|
172
|
-
| Knowledge fragments | 45 (testing, stack patterns) |
|
|
173
|
-
| Agent memory sidecars | 9 |
|
|
174
|
-
| Output artifact dirs | 4 |
|
|
150
|
+
Workflows support three modes: **normal** (pauses for confirmation at each checkpoint), **YOLO** (auto-proceeds without pausing), and **planning** (presents a structured execution plan for approval before any steps run). You can switch between normal and YOLO mid-workflow.
|
|
175
151
|
|
|
176
152
|
---
|
|
177
153
|
|
|
178
154
|
## Agents
|
|
179
155
|
|
|
180
|
-
Every agent has a name, persona, and specialization.
|
|
181
|
-
|
|
182
|
-
### Orchestrator
|
|
183
|
-
|
|
184
|
-
| Agent | Name | Role |
|
|
185
|
-
|-------|------|------|
|
|
186
|
-
| Orchestrator | Gaia | Routes requests, manages resources, presents categorized menus |
|
|
156
|
+
Every agent has a name, persona, and specialization. Activate any agent directly with `/gaia-agent-{name}` or let the orchestrator route you.
|
|
187
157
|
|
|
188
158
|
### Lifecycle Agents
|
|
189
159
|
|
|
190
160
|
| Agent | Name | Specialization | Command |
|
|
191
161
|
|-------|------|---------------|---------|
|
|
162
|
+
| Orchestrator | Gaia | Routes requests, manages resources | `/gaia` |
|
|
192
163
|
| Business Analyst | Elena | Market research, domain analysis, product briefs | `/gaia-agent-analyst` |
|
|
193
164
|
| Product Manager | Derek | PRDs, requirements, stakeholder alignment | `/gaia-agent-pm` |
|
|
194
165
|
| UX Designer | Christy | User research, interaction design, UI patterns | `/gaia-agent-ux-designer` |
|
|
@@ -237,10 +208,6 @@ All developer agents extend a shared base with common dev patterns. Each adds st
|
|
|
237
208
|
|
|
238
209
|
Workflows are structured multi-step processes. Each has a `workflow.yaml` config, `instructions.xml` with step-by-step guidance, and a `checklist.md` for quality gates.
|
|
239
210
|
|
|
240
|
-
### Model Assignment
|
|
241
|
-
|
|
242
|
-
Each command specifies which Claude model to use. **Opus** is assigned to workflows requiring deep reasoning, multi-step synthesis, or complex analysis. **Sonnet** is assigned to structured generation, template-following, and status reporting tasks.
|
|
243
|
-
|
|
244
211
|
### Phase 1: Analysis
|
|
245
212
|
|
|
246
213
|
| Command | Workflow | Agent | Model | Output |
|
|
@@ -302,8 +269,6 @@ Each command specifies which Claude model to use. **Opus** is assigned to workfl
|
|
|
302
269
|
|
|
303
270
|
### Quick Flow
|
|
304
271
|
|
|
305
|
-
Fast-track workflows for small changes that skip the full lifecycle ceremony.
|
|
306
|
-
|
|
307
272
|
| Command | Workflow | Model | Description |
|
|
308
273
|
|---------|----------|-------|-------------|
|
|
309
274
|
| `/gaia-quick-spec` | Quick Spec | Sonnet | Rapid tech spec — skip full PRD |
|
|
@@ -323,16 +288,7 @@ Fast-track workflows for small changes that skip the full lifecycle ceremony.
|
|
|
323
288
|
|
|
324
289
|
### Testing Workflows
|
|
325
290
|
|
|
326
|
-
Testing workflows are **integrated into the main lifecycle** — they are not optional standalone tools.
|
|
327
|
-
|
|
328
|
-
| Lifecycle Point | Required Testing Workflow | Gate Type |
|
|
329
|
-
|---|---|---|
|
|
330
|
-
| After `/gaia-create-arch` | `/gaia-test-design` | HALT at `/gaia-create-epics` |
|
|
331
|
-
| After `/gaia-test-design` | `/gaia-test-framework` (optional) | Prompted if no framework detected |
|
|
332
|
-
| Before `/gaia-dev-story` (high-risk) | `/gaia-atdd` | HALT (conditional on risk_level) |
|
|
333
|
-
| Before `/gaia-readiness-check` | `/gaia-trace` + `/gaia-ci-setup` | HALT |
|
|
334
|
-
| During `/gaia-brownfield` | `/gaia-nfr` + `/gaia-perf-testing` | REQUIRED steps |
|
|
335
|
-
| Before `/gaia-deploy-checklist` | traceability + CI + readiness PASS | HALT |
|
|
291
|
+
Testing workflows are **integrated into the main lifecycle** — they are not optional standalone tools.
|
|
336
292
|
|
|
337
293
|
| Command | Workflow | Agent | Model | Output |
|
|
338
294
|
|---------|----------|-------|-------|--------|
|
|
@@ -351,8 +307,6 @@ Testing workflows are **integrated into the main lifecycle** — they are not op
|
|
|
351
307
|
|
|
352
308
|
### Anytime Workflows
|
|
353
309
|
|
|
354
|
-
Available at any point in the lifecycle.
|
|
355
|
-
|
|
356
310
|
| Command | Workflow | Model | Description |
|
|
357
311
|
|---------|----------|-------|-------------|
|
|
358
312
|
| `/gaia-brownfield` | Brownfield Onboarding | Opus | 6-step orchestration with subagents for existing projects |
|
|
@@ -362,417 +316,158 @@ Available at any point in the lifecycle.
|
|
|
362
316
|
| `/gaia-brainstorming` | Brainstorming | Sonnet | Facilitated brainstorming session |
|
|
363
317
|
| `/gaia-party` | Party Mode | Sonnet | Multi-agent group discussion |
|
|
364
318
|
| `/gaia-advanced-elicitation` | Advanced Elicitation | Opus | Deep requirements elicitation |
|
|
365
|
-
| `/gaia-memory-hygiene` | Memory Hygiene | Sonnet | Detect stale
|
|
366
|
-
|
|
367
|
-
### Brownfield Onboarding (Deep Dive)
|
|
368
|
-
|
|
369
|
-
The brownfield onboarding workflow (`/gaia-brownfield`) is a comprehensive 9-step process that transforms an existing codebase into a fully documented developer knowledge base. It detects project capabilities during scanning and conditionally generates specialized documentation.
|
|
370
|
-
|
|
371
|
-
**Step 1 — Deep Project Discovery** scans the codebase and sets capability flags:
|
|
372
|
-
|
|
373
|
-
| Flag | Detection |
|
|
374
|
-
|------|-----------|
|
|
375
|
-
| `{has_apis}` | Route definitions, controllers, OpenAPI/Swagger specs |
|
|
376
|
-
| `{has_events}` | Kafka, RabbitMQ, SNS-SQS, Redis pub-sub, NATS patterns |
|
|
377
|
-
| `{has_external_deps}` | Outbound HTTP clients, SDKs, service URLs |
|
|
378
|
-
| `{has_frontend}` | React, Angular, Vue, Flutter, SwiftUI, CSS frameworks |
|
|
379
|
-
|
|
380
|
-
**Steps 2–4** are conditional — they only run when the corresponding capability is detected:
|
|
381
|
-
|
|
382
|
-
| Step | Condition | Output |
|
|
383
|
-
|------|-----------|--------|
|
|
384
|
-
| 2. API Documentation | `{has_apis}` | OpenAPI 3.x spec (discovered or generated), endpoint inventory, Mermaid flow diagram |
|
|
385
|
-
| 3. UX Design Assessment | `{has_frontend}` | UI patterns, Mermaid navigation sitemap, accessibility assessment, UX gaps |
|
|
386
|
-
| 4. Event & Messaging Catalog | `{has_events}` | Producer/consumer tables, delivery guarantees, Mermaid event flow diagrams |
|
|
387
|
-
|
|
388
|
-
**Steps 5–10** always run:
|
|
389
|
-
|
|
390
|
-
| Step | Output |
|
|
391
|
-
|------|--------|
|
|
392
|
-
| 5. Dependency Map | External services, infrastructure, library deps with CVE risk, Mermaid dependency graph |
|
|
393
|
-
| 6. NFR Assessment & Baselines | Code quality, security posture, performance, test coverage — output to `test-artifacts/nfr-assessment.md` (canonical) |
|
|
394
|
-
| 7. Performance Test Plan | Performance budgets, load test scenarios, Core Web Vitals targets — output to `test-artifacts/` (**REQUIRED**) |
|
|
395
|
-
| 8. Create PRD for Gaps | Gap-focused PRD with NFR baselines from step 6, references all upstream artifacts |
|
|
396
|
-
| 9. Map Architecture with Diagrams | C4 Mermaid diagrams (Level 1 + 2), 3–5 sequence diagrams, as-is/target delta |
|
|
397
|
-
| 10. Epics/Stories & Onboard Developer | Gap stories + developer knowledge base index linking all artifacts |
|
|
398
|
-
|
|
399
|
-
**Design rules:**
|
|
400
|
-
- All diagrams use **Mermaid syntax** — no ASCII art
|
|
401
|
-
- All API docs use **Swagger/OpenAPI format**
|
|
402
|
-
- Flow diagrams limited to **3–5 key flows** per document
|
|
403
|
-
- NFR baselines are **measured from the codebase**, not estimated
|
|
404
|
-
|
|
405
|
-
**Output artifacts** (up to 11, depending on detected capabilities):
|
|
406
|
-
|
|
407
|
-
| Artifact | Location | Always | Conditional |
|
|
408
|
-
|----------|----------|--------|-------------|
|
|
409
|
-
| `project-documentation.md` | `planning-artifacts/` | Yes | — |
|
|
410
|
-
| `api-documentation.md` | `planning-artifacts/` | — | `{has_apis}` |
|
|
411
|
-
| `ux-design.md` | `planning-artifacts/` | — | `{has_frontend}` |
|
|
412
|
-
| `event-catalog.md` | `planning-artifacts/` | — | `{has_events}` |
|
|
413
|
-
| `dependency-map.md` | `planning-artifacts/` | Yes | — |
|
|
414
|
-
| `nfr-assessment.md` | `test-artifacts/` | Yes | — |
|
|
415
|
-
| `performance-test-plan-{date}.md` | `test-artifacts/` | Yes | — |
|
|
416
|
-
| `prd.md` | `planning-artifacts/` | Yes | — |
|
|
417
|
-
| `architecture.md` | `planning-artifacts/` | Yes | — |
|
|
418
|
-
| `epics-and-stories.md` | `planning-artifacts/` | Yes | — |
|
|
419
|
-
| `brownfield-onboarding.md` | `planning-artifacts/` | Yes | — |
|
|
420
|
-
|
|
421
|
-
---
|
|
422
|
-
|
|
423
|
-
## Review & Utility Tasks
|
|
424
|
-
|
|
425
|
-
Standalone tasks that can be run anytime without a full workflow. These are single-step operations for reviews, audits, and document utilities.
|
|
319
|
+
| `/gaia-memory-hygiene` | Memory Hygiene | Sonnet | Detect stale decisions in agent memory |
|
|
426
320
|
|
|
427
|
-
###
|
|
321
|
+
### Review & Utility Tasks
|
|
428
322
|
|
|
429
323
|
| Command | Task | Model | Description |
|
|
430
324
|
|---------|------|-------|-------------|
|
|
431
|
-
| `/gaia-adversarial` | Adversarial Review | Opus | Cynical critical review — finds weaknesses |
|
|
325
|
+
| `/gaia-adversarial` | Adversarial Review | Opus | Cynical critical review from 10 perspectives — finds weaknesses |
|
|
432
326
|
| `/gaia-edge-cases` | Edge Case Hunter | Opus | Identify edge cases and boundary conditions |
|
|
433
327
|
| `/gaia-review-security` | Security Review | Opus | OWASP-focused security review |
|
|
434
328
|
| `/gaia-review-api` | API Design Review | Opus | Review REST API against standards |
|
|
435
329
|
| `/gaia-review-deps` | Dependency Audit | Sonnet | Scan dependencies for vulnerabilities |
|
|
436
330
|
| `/gaia-review-a11y` | Accessibility Review | Sonnet | WCAG 2.1 compliance review |
|
|
437
331
|
| `/gaia-review-perf` | Performance Review | Opus | Code-level performance review |
|
|
332
|
+
| `/gaia-editorial-prose` | Editorial Prose | Sonnet | Clinical copy-editing review |
|
|
333
|
+
| `/gaia-editorial-structure` | Editorial Structure | Sonnet | Structural editing review |
|
|
334
|
+
| `/gaia-summarize` | Summarize Document | Sonnet | Generate executive summary |
|
|
335
|
+
| `/gaia-index-docs` | Index Docs | Sonnet | Generate document index for a folder |
|
|
336
|
+
| `/gaia-shard-doc` | Shard Document | Sonnet | Split large docs into sections |
|
|
337
|
+
| `/gaia-merge-docs` | Merge Documents | Sonnet | Merge multiple markdown files |
|
|
338
|
+
| `/gaia-changelog` | Generate Changelog | Sonnet | Changelog from git history |
|
|
438
339
|
|
|
439
|
-
###
|
|
440
|
-
|
|
441
|
-
The adversarial review is a cynical, critical examination designed to find flaws, gaps, and weaknesses in any document or design. It deliberately assumes nothing works as claimed and attacks from **10 different perspectives**:
|
|
442
|
-
|
|
443
|
-
| Perspective | What it challenges |
|
|
444
|
-
|---|---|
|
|
445
|
-
| **Feasibility** | Can this actually be built as described? |
|
|
446
|
-
| **Completeness** | What's missing that should be there? |
|
|
447
|
-
| **Contradictions** | Do any sections contradict each other? |
|
|
448
|
-
| **Assumptions** | What unstated assumptions could be wrong? |
|
|
449
|
-
| **Scale** | Will this work at 10x/100x expected load? |
|
|
450
|
-
| **Failure modes** | What happens when things go wrong? |
|
|
451
|
-
| **Dependencies** | What external factors could break this? |
|
|
452
|
-
| **Security** | What attack surfaces are exposed? |
|
|
453
|
-
| **User impact** | Where will users get confused or frustrated? |
|
|
454
|
-
| **Business risk** | What could make this commercially unviable? |
|
|
455
|
-
|
|
456
|
-
**Output:** A ranked findings report with severity (critical/high/medium/low), confidence (certain/likely/possible), a threat summary of the top 3 issues, and an overall risk assessment. The review only identifies problems — it does not suggest fixes.
|
|
457
|
-
|
|
458
|
-
**When to use it:** The adversarial review is integrated as an optional prompt at 4 key lifecycle points:
|
|
459
|
-
|
|
460
|
-
| Trigger Point | When Prompted | Recommended For |
|
|
461
|
-
|---|---|---|
|
|
462
|
-
| After `/gaia-create-prd` | End of PRD creation (step 11) | Complex or high-stakes products |
|
|
463
|
-
| After `/gaia-create-arch` | End of architecture design (step 9) | Distributed systems, high-scale projects |
|
|
464
|
-
| After `/gaia-create-epics` | End of epic/story creation (step 8) | Before sprint planning |
|
|
465
|
-
| After `/gaia-readiness-check` | End of readiness check (step 9) | Strongly recommended — last chance before build |
|
|
466
|
-
|
|
467
|
-
At each point you can accept or skip. It can also be run independently anytime with `/gaia-adversarial` on any artifact.
|
|
468
|
-
|
|
469
|
-
### Lifecycle-Integrated Reviews
|
|
470
|
-
|
|
471
|
-
In addition to the adversarial review, several other tasks are integrated into the lifecycle at natural trigger points:
|
|
472
|
-
|
|
473
|
-
**Review Gate members** (required — story cannot move to `done` without all 6 passing):
|
|
474
|
-
|
|
475
|
-
| Review Command | Gate Row | Verdict | What It Checks |
|
|
476
|
-
|---|---|---|---|
|
|
477
|
-
| `/gaia-code-review` | Code Review | APPROVE / REQUEST_CHANGES | Correctness, security, performance, readability |
|
|
478
|
-
| `/gaia-qa-tests` | QA Tests | PASSED / FAILED | E2E and API test generation + execution |
|
|
479
|
-
| `/gaia-security-review` | Security Review | PASSED / FAILED | OWASP Top 10, secrets, auth patterns |
|
|
480
|
-
| `/gaia-test-automate` | Test Automation | PASSED / FAILED | Coverage gaps, uncovered code paths |
|
|
481
|
-
| `/gaia-test-review` | Test Review | PASSED / FAILED | Test quality, flakiness, isolation, determinism |
|
|
482
|
-
| `/gaia-review-perf` | Performance Review | PASSED / FAILED | N+1 queries, memory, bundle size, caching, complexity. Auto-passes if no performance-relevant code changes |
|
|
483
|
-
|
|
484
|
-
**Optional prompts** (offered at natural lifecycle points — user can accept or skip):
|
|
485
|
-
|
|
486
|
-
| Review Command | Trigger Point | When Prompted | Recommended For |
|
|
487
|
-
|---|---|---|---|
|
|
488
|
-
| `/gaia-review-api` | After `/gaia-create-arch` | End of architecture design (step 9) | Projects with REST APIs |
|
|
489
|
-
| `/gaia-review-a11y` | After `/gaia-create-ux` | End of UX design (step 8) | User-facing applications |
|
|
490
|
-
| `/gaia-edge-cases` | After `/gaia-create-epics` | End of epic/story creation (step 8) | Before sprint planning |
|
|
491
|
-
| `/gaia-review-deps` | During `/gaia-brownfield` | After dependency map (step 5) | Brownfield projects |
|
|
492
|
-
| `/gaia-review-perf` | Review Gate member | Required (auto-passes if no perf-relevant changes) | See Review Gate members table above |
|
|
493
|
-
| `/gaia-a11y-testing` | During `/gaia-deploy-checklist` | After loading context (step 2) | Frontend/user-facing applications |
|
|
494
|
-
| `/gaia-mobile-testing` | During `/gaia-sprint-plan` | After story selection (step 4) | Mobile/responsive applications |
|
|
495
|
-
|
|
496
|
-
All reviews can also be run independently anytime on any artifact or codebase.
|
|
497
|
-
|
|
498
|
-
### Editorial & Documentation
|
|
499
|
-
|
|
500
|
-
| Command | Task | Description |
|
|
501
|
-
|---------|------|-------------|
|
|
502
|
-
| `/gaia-editorial-prose` | Editorial Prose | Clinical copy-editing review |
|
|
503
|
-
| `/gaia-editorial-structure` | Editorial Structure | Structural editing review |
|
|
504
|
-
| `/gaia-summarize` | Summarize Document | Generate executive summary |
|
|
505
|
-
| `/gaia-index-docs` | Index Docs | Generate document index for a folder |
|
|
506
|
-
| `/gaia-shard-doc` | Shard Document | Split large docs into sections |
|
|
507
|
-
| `/gaia-merge-docs` | Merge Documents | Merge multiple markdown files |
|
|
508
|
-
| `/gaia-changelog` | Generate Changelog | Changelog from git history |
|
|
509
|
-
|
|
510
|
-
### Framework
|
|
511
|
-
|
|
512
|
-
| Command | Task | Description |
|
|
513
|
-
|---------|------|-------------|
|
|
514
|
-
| `/gaia-build-configs` | Build Configs | Regenerate pre-resolved config files |
|
|
515
|
-
| `/gaia-validate-framework` | Validate Framework | Self-validation and consistency check |
|
|
516
|
-
| `/gaia-resume` | Resume | Resume from last checkpoint after context loss |
|
|
517
|
-
|
|
518
|
-
---
|
|
519
|
-
|
|
520
|
-
## Shared Skills
|
|
521
|
-
|
|
522
|
-
Skills are loaded just-in-time by developer agents. Each skill is divided into sections so only the relevant portion is loaded (keeping context usage low).
|
|
523
|
-
|
|
524
|
-
| Skill | Sections | Used by |
|
|
525
|
-
|-------|----------|---------|
|
|
526
|
-
| **Git Workflow** | branching, commits, pull-requests, conflict-resolution | All devs, Scrum Master, DevOps |
|
|
527
|
-
| **API Design** | rest-conventions, graphql, openapi, versioning, error-standards | All devs, Architect, Data Engineer |
|
|
528
|
-
| **Database Design** | schema-design, migrations, indexing, orm-patterns | Java, Python, Architect, Data Engineer |
|
|
529
|
-
| **Docker Workflow** | multi-stage-builds, compose, security-scanning | TS, Angular, Java, Python, Mobile, DevOps |
|
|
530
|
-
| **Testing Patterns** | tdd-cycle, unit-testing, integration-testing, test-doubles | All devs, QA, Test Architect |
|
|
531
|
-
| **Code Review Standards** | review-checklist, solid-principles, complexity-metrics | All devs |
|
|
532
|
-
| **Documentation Standards** | readme-template, adr-format, inline-comments, api-docs | All devs, Tech Writer, Analyst, PM |
|
|
533
|
-
| **Security Basics** | owasp-top-10, input-validation, secrets-management, cors-csrf | All devs, Architect, Security, DevOps |
|
|
534
|
-
|
|
535
|
-
---
|
|
536
|
-
|
|
537
|
-
## Knowledge Fragments
|
|
538
|
-
|
|
539
|
-
Stack-specific and domain-specific knowledge loaded on demand.
|
|
540
|
-
|
|
541
|
-
### Developer Knowledge (by stack)
|
|
542
|
-
|
|
543
|
-
| Stack | Fragments |
|
|
544
|
-
|-------|-----------|
|
|
545
|
-
| TypeScript | React patterns, Next.js patterns, Express patterns, TS conventions |
|
|
546
|
-
| Angular | Angular conventions, Angular patterns, NgRx state, RxJS patterns |
|
|
547
|
-
| Flutter | Dart conventions, Widget patterns, State management, Platform channels |
|
|
548
|
-
| Java | Spring Boot patterns, JPA patterns, Microservices, Maven/Gradle |
|
|
549
|
-
| Python | Python conventions, Django patterns, FastAPI patterns, Data pipelines |
|
|
550
|
-
| Mobile | React Native patterns, Swift patterns, Kotlin patterns, Mobile testing |
|
|
551
|
-
|
|
552
|
-
### Testing Knowledge (by tier)
|
|
340
|
+
### Framework Utilities
|
|
553
341
|
|
|
554
|
-
|
|
|
555
|
-
|
|
556
|
-
|
|
|
557
|
-
|
|
|
558
|
-
|
|
|
559
|
-
|
|
|
560
|
-
| **Accessibility** | WCAG checks, Axe-core patterns |
|
|
561
|
-
| **Mobile** | Appium patterns, React Native testing, Responsive testing |
|
|
562
|
-
| **Unit Testing** | Jest/Vitest patterns, JUnit5 patterns, Pytest patterns |
|
|
342
|
+
| Command | Task | Model | Description |
|
|
343
|
+
|---------|------|-------|-------------|
|
|
344
|
+
| `/gaia-build-configs` | Build Configs | Sonnet | Regenerate pre-resolved config files |
|
|
345
|
+
| `/gaia-validate-framework` | Validate Framework | Sonnet | Self-validation and consistency check |
|
|
346
|
+
| `/gaia-resume` | Resume | Sonnet | Resume from last checkpoint after context loss |
|
|
347
|
+
| `/gaia-run-all-reviews` | Run All Reviews | Sonnet | Execute all 6 review gate checks sequentially |
|
|
563
348
|
|
|
564
349
|
---
|
|
565
350
|
|
|
566
|
-
##
|
|
351
|
+
## Typical Workflow
|
|
567
352
|
|
|
568
|
-
|
|
353
|
+
### Greenfield — new project from idea to deployment
|
|
569
354
|
|
|
570
355
|
```
|
|
571
|
-
|
|
572
|
-
|
|
573
|
-
|
|
574
|
-
├── test-artifacts/ # Test plans, traceability, accessibility
|
|
575
|
-
└── creative-artifacts/ # Design thinking, innovation, pitch decks
|
|
576
|
-
```
|
|
577
|
-
|
|
578
|
-
---
|
|
356
|
+
# Phase 1: Analysis
|
|
357
|
+
/gaia-brainstorm → brainstorm the idea
|
|
358
|
+
/gaia-product-brief → create a product brief
|
|
579
359
|
|
|
580
|
-
|
|
581
|
-
|
|
582
|
-
|
|
583
|
-
|
|
584
|
-
| Template | Used By | Description |
|
|
585
|
-
|----------|---------|-------------|
|
|
586
|
-
| `prd-template.md` | create-prd | Product requirements document |
|
|
587
|
-
| `architecture-template.md` | create-architecture | Greenfield architecture document |
|
|
588
|
-
| `brownfield-architecture-template.md` | brownfield-onboarding | As-is/target architecture with Mermaid C4 diagrams |
|
|
589
|
-
| `story-template.md` | create-epics-stories | User story with acceptance criteria |
|
|
590
|
-
| `sprint-plan-template.md` | sprint-planning | Sprint plan with capacity and goals |
|
|
591
|
-
| `product-brief-template.md` | create-product-brief | Product brief for stakeholder alignment |
|
|
592
|
-
| `review-template.md` | code-review | Code review report |
|
|
593
|
-
| `test-plan-template.md` | test-design | Test plan with strategy and coverage |
|
|
594
|
-
| `deployment-template.md` | release-plan | Deployment plan and checklist |
|
|
595
|
-
| `api-documentation-template.md` | brownfield-onboarding | OpenAPI/Swagger API documentation |
|
|
596
|
-
| `event-catalog-template.md` | brownfield-onboarding | Event producer/consumer catalog |
|
|
597
|
-
| `dependency-map-template.md` | brownfield-onboarding | Service, infrastructure, and library dependencies |
|
|
598
|
-
| `nfr-assessment-template.md` | brownfield-onboarding | Non-functional requirements baselines |
|
|
599
|
-
| `ux-design-assessment-template.md` | brownfield-onboarding | UI patterns and accessibility assessment |
|
|
600
|
-
| `epic-status-template.md` | epic-status | Epic completion dashboard with progress bars |
|
|
601
|
-
| `brownfield-onboarding-template.md` | brownfield-onboarding | Developer knowledge base index |
|
|
360
|
+
# Phase 2: Planning
|
|
361
|
+
/gaia-create-prd → write the PRD
|
|
362
|
+
/gaia-create-ux → design the UX
|
|
602
363
|
|
|
603
|
-
|
|
364
|
+
# Phase 3: Solutioning
|
|
365
|
+
/gaia-create-arch → design the architecture
|
|
366
|
+
/gaia-test-design → create test plan
|
|
367
|
+
/gaia-create-epics → break into epics and stories
|
|
368
|
+
/gaia-trace → generate traceability matrix
|
|
369
|
+
/gaia-ci-setup → scaffold CI pipeline
|
|
370
|
+
/gaia-readiness-check → verify everything is ready
|
|
604
371
|
|
|
605
|
-
|
|
372
|
+
# Phase 4: Implementation (repeat per sprint)
|
|
373
|
+
/gaia-sprint-plan → plan the sprint
|
|
374
|
+
/gaia-dev-story → implement stories
|
|
375
|
+
/gaia-run-all-reviews → run all 6 review gates
|
|
376
|
+
/gaia-retro → sprint retrospective
|
|
606
377
|
|
|
607
|
-
|
|
378
|
+
# Phase 5: Deployment
|
|
379
|
+
/gaia-deploy-checklist → pre-deploy verification
|
|
380
|
+
/gaia-post-deploy → post-deploy health check
|
|
381
|
+
```
|
|
608
382
|
|
|
609
|
-
|
|
383
|
+
For **brownfield projects** (existing codebases), start with `/gaia-brownfield` — a 6-step orchestration that scans your codebase, generates documentation (APIs, UX, events, dependencies, NFR baselines), creates a gap-focused PRD, and maps the architecture. Then continue from Phase 3 above.
|
|
610
384
|
|
|
611
385
|
---
|
|
612
386
|
|
|
613
|
-
##
|
|
614
|
-
|
|
615
|
-
Each agent has a persistent memory sidecar that survives across sessions:
|
|
387
|
+
## Architecture
|
|
616
388
|
|
|
617
389
|
```
|
|
618
|
-
_gaia/
|
|
619
|
-
├──
|
|
620
|
-
├──
|
|
621
|
-
|
|
622
|
-
├──
|
|
623
|
-
├──
|
|
624
|
-
|
|
625
|
-
├──
|
|
626
|
-
|
|
627
|
-
|
|
390
|
+
_gaia/
|
|
391
|
+
├── _config/ # Global config, manifests
|
|
392
|
+
│ ├── global.yaml # Project settings — single source of truth
|
|
393
|
+
│ └── manifest.yaml # Module versions
|
|
394
|
+
├── _memory/ # Persistent agent memory + checkpoints
|
|
395
|
+
│ ├── checkpoints/ # Workflow progress snapshots (sha256-verified)
|
|
396
|
+
│ └── *-sidecar/ # Per-agent persistent memory (9 sidecars)
|
|
397
|
+
├── core/ # Execution engine, protocols, shared tasks
|
|
398
|
+
│ └── engine/ # workflow.xml (7-step flow), task-runner.xml
|
|
399
|
+
├── lifecycle/ # 5 phases: analysis → deployment
|
|
400
|
+
│ ├── agents/ # 11 lifecycle agents
|
|
401
|
+
│ ├── workflows/ # 36 workflows across 5 phases
|
|
402
|
+
│ └── templates/ # 18 document templates
|
|
403
|
+
├── dev/ # Developer tooling
|
|
404
|
+
│ ├── agents/ # 6 stack-specific developers
|
|
405
|
+
│ ├── skills/ # 8 shared skills (sectioned loading)
|
|
406
|
+
│ └── knowledge/ # Stack-specific patterns
|
|
407
|
+
├── creative/ # 6 creative agents + 7 workflows
|
|
408
|
+
└── testing/ # Test Architect + 12 testing workflows
|
|
628
409
|
```
|
|
629
410
|
|
|
630
|
-
|
|
411
|
+
### At a glance
|
|
412
|
+
|
|
413
|
+
| Component | Count |
|
|
414
|
+
|-----------|-------|
|
|
415
|
+
| Agents | 25 with distinct personas |
|
|
416
|
+
| Workflows | 64 across 5 lifecycle phases |
|
|
417
|
+
| Standalone tasks | 15 (reviews, audits, utilities) |
|
|
418
|
+
| Slash commands | 104 |
|
|
419
|
+
| Shared skills | 8 with 47 loadable sections |
|
|
420
|
+
| Knowledge fragments | 45 |
|
|
421
|
+
| Document templates | 18 |
|
|
422
|
+
| Quality gates | 17 (enforced, not advisory) |
|
|
631
423
|
|
|
632
424
|
---
|
|
633
425
|
|
|
634
426
|
## Configuration
|
|
635
427
|
|
|
636
|
-
|
|
637
|
-
|
|
638
|
-
The single source of truth for project settings at `_gaia/_config/global.yaml`:
|
|
428
|
+
The single source of truth is `_gaia/_config/global.yaml`:
|
|
639
429
|
|
|
640
430
|
```yaml
|
|
641
431
|
framework_name: "GAIA"
|
|
642
|
-
framework_version: "1.27.
|
|
643
|
-
|
|
432
|
+
framework_version: "1.27.2"
|
|
644
433
|
user_name: "your-name"
|
|
645
434
|
project_name: "your-project"
|
|
646
|
-
project_root: "{project-root}"
|
|
647
|
-
|
|
648
|
-
output_folder: "{project-root}/docs"
|
|
649
|
-
planning_artifacts: "{project-root}/docs/planning-artifacts"
|
|
650
|
-
implementation_artifacts: "{project-root}/docs/implementation-artifacts"
|
|
651
|
-
test_artifacts: "{project-root}/docs/test-artifacts"
|
|
652
|
-
creative_artifacts: "{project-root}/docs/creative-artifacts"
|
|
653
435
|
```
|
|
654
436
|
|
|
655
|
-
|
|
656
|
-
|
|
657
|
-
### Pre-resolved configs
|
|
658
|
-
|
|
659
|
-
Each module (core, lifecycle, creative, testing) has a `.resolved/` directory for pre-built config files. These eliminate runtime config resolution overhead. Generate them with `/gaia-build-configs`.
|
|
437
|
+
After changing `global.yaml`, run `/gaia-build-configs` to regenerate pre-resolved configs. Each module has a `.resolved/` directory that eliminates runtime config resolution overhead.
|
|
660
438
|
|
|
661
439
|
---
|
|
662
440
|
|
|
663
|
-
##
|
|
664
|
-
|
|
665
|
-
Pre-built team compositions for different project types:
|
|
441
|
+
## Checkpoint & Resume
|
|
666
442
|
|
|
667
|
-
|
|
668
|
-
|------|-------|--------|
|
|
669
|
-
| Full | Complete coverage | All lifecycle + dev agents |
|
|
670
|
-
| Planning | Requirements & design | Analyst, PM, UX Designer, Architect |
|
|
671
|
-
| Implementation | Build & ship | SM, dev agents, QA, DevOps |
|
|
672
|
-
| Quick Ship | Minimal ceremony | PM, dev agent, QA |
|
|
673
|
-
| Enterprise | Governance-heavy | Full team + Security, Performance |
|
|
674
|
-
| Security-Focused | Security-first | Architect, Security, DevOps, QA |
|
|
675
|
-
| Data-Intensive | Data pipelines | Architect, Data Engineer, Python Dev |
|
|
443
|
+
Long-running workflows save checkpoints to `_gaia/_memory/checkpoints/` with sha256 checksums of all files touched. If your session is interrupted, run `/gaia-resume` — it validates file integrity before resuming from the last completed step.
|
|
676
444
|
|
|
677
445
|
---
|
|
678
446
|
|
|
679
|
-
##
|
|
680
|
-
|
|
681
|
-
### Greenfield — New project from idea to deployment
|
|
682
|
-
|
|
683
|
-
```
|
|
684
|
-
# Phase 1: Analysis
|
|
685
|
-
/gaia-brainstorm → brainstorm the idea
|
|
686
|
-
/gaia-product-brief → create a product brief
|
|
687
|
-
/gaia-market-research → validate market fit
|
|
688
|
-
|
|
689
|
-
# Phase 2: Planning
|
|
690
|
-
/gaia-create-prd → write the PRD (optional: adversarial review)
|
|
691
|
-
/gaia-create-ux → design the UX (optional: accessibility review)
|
|
692
|
-
|
|
693
|
-
# Phase 3: Solutioning
|
|
694
|
-
/gaia-create-arch → design the architecture (optional: API review, adversarial review)
|
|
695
|
-
/gaia-test-design → create test plan (optional: scaffold test framework)
|
|
696
|
-
/gaia-create-epics → break into epics and stories (optional: edge cases, adversarial review)
|
|
697
|
-
/gaia-trace → generate traceability matrix
|
|
698
|
-
/gaia-ci-setup → scaffold CI pipeline
|
|
699
|
-
/gaia-readiness-check → verify everything is ready (optional: adversarial review)
|
|
447
|
+
## Agent Memory
|
|
700
448
|
|
|
701
|
-
|
|
702
|
-
/gaia-sprint-plan → plan the sprint (optional: mobile testing)
|
|
703
|
-
/gaia-create-story → create detailed stories
|
|
704
|
-
/gaia-validate-story → validate story completeness
|
|
705
|
-
/gaia-fix-story → fix issues from validation
|
|
706
|
-
/gaia-atdd → write acceptance tests (REQUIRED for high-risk stories)
|
|
707
|
-
/gaia-dev-story → implement stories
|
|
708
|
-
/gaia-code-review → review the code ─┐
|
|
709
|
-
/gaia-qa-tests → generate tests │
|
|
710
|
-
/gaia-security-review → security audit │ Review Gate
|
|
711
|
-
/gaia-test-automate → expand test coverage │ (all 6 must
|
|
712
|
-
/gaia-test-review → review test quality │ PASS before
|
|
713
|
-
/gaia-review-perf → performance review ─┘ story → done)
|
|
714
|
-
/gaia-triage-findings → triage dev findings into backlog
|
|
715
|
-
/gaia-retro → sprint retrospective
|
|
449
|
+
Each agent has a persistent memory sidecar (`_gaia/_memory/*-sidecar/`) that stores decisions, patterns, and context across sessions. Agents become more effective the more you use them. Run `/gaia-memory-hygiene` periodically to detect stale or contradicted decisions.
|
|
716
450
|
|
|
717
|
-
|
|
718
|
-
/gaia-release-plan → plan the release
|
|
719
|
-
/gaia-deploy-checklist → pre-deploy verification (optional: a11y testing)
|
|
720
|
-
/gaia-post-deploy → post-deploy health check
|
|
721
|
-
```
|
|
451
|
+
---
|
|
722
452
|
|
|
723
|
-
|
|
453
|
+
## Limitations
|
|
724
454
|
|
|
725
|
-
|
|
726
|
-
|
|
727
|
-
|
|
728
|
-
1. Scan codebase → project-documentation.md
|
|
729
|
-
2. API documentation (if APIs detected)
|
|
730
|
-
3. UX assessment (if frontend detected)
|
|
731
|
-
4. Event catalog (if messaging detected)
|
|
732
|
-
5. Dependency map
|
|
733
|
-
6. NFR assessment & baselines
|
|
734
|
-
7. Performance test plan
|
|
735
|
-
8. Gap-focused PRD (gaps only, not existing features)
|
|
736
|
-
9. Architecture with as-is/target diagrams
|
|
737
|
-
10. Epics/stories for gaps + developer knowledge base
|
|
738
|
-
|
|
739
|
-
# Phase 3: Readiness
|
|
740
|
-
/gaia-test-design → create test plan (optional: scaffold test framework)
|
|
741
|
-
/gaia-trace → generate traceability matrix
|
|
742
|
-
/gaia-ci-setup → scaffold CI pipeline
|
|
743
|
-
/gaia-readiness-check → verify everything is ready (optional: adversarial review)
|
|
455
|
+
- **Single-user only** — GAIA uses markdown files for state management (stories, sprint status, architecture docs). Multiple team members editing the same project will run into file conflicts.
|
|
456
|
+
- **Claude Code required** — GAIA is built specifically for Claude Code and cannot run on other AI coding assistants.
|
|
457
|
+
- **Context budget** — Complex workflows can consume significant context. The framework enforces a 40K token budget per activation with just-in-time loading to manage this, but very large projects may hit limits.
|
|
744
458
|
|
|
745
|
-
|
|
746
|
-
/gaia-sprint-plan → plan the sprint (optional: mobile testing)
|
|
747
|
-
/gaia-create-story → create detailed stories
|
|
748
|
-
/gaia-validate-story → validate story completeness
|
|
749
|
-
/gaia-fix-story → fix issues from validation
|
|
750
|
-
/gaia-atdd → write acceptance tests (REQUIRED for high-risk stories)
|
|
751
|
-
/gaia-dev-story → implement gap stories
|
|
752
|
-
/gaia-code-review → review the code ─┐
|
|
753
|
-
/gaia-qa-tests → generate tests │
|
|
754
|
-
/gaia-security-review → security audit │ Review Gate
|
|
755
|
-
/gaia-test-automate → expand test coverage │ (all 6 must
|
|
756
|
-
/gaia-test-review → review test quality │ PASS before
|
|
757
|
-
/gaia-review-perf → performance review ─┘ story → done)
|
|
758
|
-
/gaia-triage-findings → triage dev findings into backlog
|
|
759
|
-
/gaia-retro → sprint retrospective
|
|
459
|
+
---
|
|
760
460
|
|
|
761
|
-
|
|
762
|
-
/gaia-release-plan → plan the release
|
|
763
|
-
/gaia-deploy-checklist → pre-deploy verification (optional: a11y testing)
|
|
764
|
-
/gaia-post-deploy → post-deploy health check
|
|
765
|
-
```
|
|
461
|
+
## Contributing
|
|
766
462
|
|
|
767
|
-
|
|
463
|
+
Contributions are welcome. Please open an issue to discuss your idea before submitting a PR.
|
|
768
464
|
|
|
769
|
-
|
|
770
|
-
/gaia-quick-spec → rapid tech spec
|
|
771
|
-
/gaia-quick-dev → implement it
|
|
772
|
-
```
|
|
465
|
+
By contributing, you agree that your contributions will be licensed under the same AGPL-3.0 license and that you grant the project maintainers the right to relicense your contributions under a commercial license (CLA).
|
|
773
466
|
|
|
774
467
|
---
|
|
775
468
|
|
|
776
469
|
## License
|
|
777
470
|
|
|
778
|
-
|
|
471
|
+
[AGPL-3.0](LICENSE)
|
|
472
|
+
|
|
473
|
+
The open-source framework is licensed under the GNU Affero General Public License v3.0.
|
package/gaia-install.sh
CHANGED
|
@@ -6,7 +6,7 @@ set -euo pipefail
|
|
|
6
6
|
# Installs, updates, validates, and reports on GAIA installations.
|
|
7
7
|
# ─────────────────────────────────────────────────────────────────────────────
|
|
8
8
|
|
|
9
|
-
readonly VERSION="1.27.
|
|
9
|
+
readonly VERSION="1.27.2"
|
|
10
10
|
readonly GITHUB_REPO="https://github.com/J-louage/Gaia-framework.git"
|
|
11
11
|
readonly MANIFEST_REL="_gaia/_config/manifest.yaml"
|
|
12
12
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "gaia-framework",
|
|
3
|
-
"version": "1.27.
|
|
3
|
+
"version": "1.27.2",
|
|
4
4
|
"description": "GAIA — Generative Agile Intelligence Architecture installer",
|
|
5
5
|
"bin": {
|
|
6
6
|
"gaia-framework": "./bin/gaia-framework.js"
|
|
@@ -18,7 +18,7 @@
|
|
|
18
18
|
"type": "git",
|
|
19
19
|
"url": "https://github.com/J-louage/Gaia-framework.git"
|
|
20
20
|
},
|
|
21
|
-
"license": "
|
|
21
|
+
"license": "AGPL-3.0",
|
|
22
22
|
"engines": {
|
|
23
23
|
"node": ">=18"
|
|
24
24
|
},
|