@jaguilar87/gaia-ops 2.2.0 → 2.2.1

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/CHANGELOG.md CHANGED
@@ -7,6 +7,68 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
7
7
 
8
8
  ---
9
9
 
10
+ ## [2.2.1] - 2025-11-10
11
+
12
+ ### Fixed - Documentation Consistency
13
+ - **README.md & README.en.md:**
14
+ - Updated version numbers from 2.1.0 → 2.2.0
15
+ - Corrected package structure (hooks/, templates/, commands/)
16
+ - Fixed hooks/ listing: now shows actual Python files (pre_tool_use.py, post_tool_use.py, etc.) instead of non-existent pre-commit
17
+ - Fixed templates/ listing: removed non-existent code-examples/, listed actual files (CLAUDE.template.md, settings.template.json)
18
+ - Added context-contracts.gcp.json and context-contracts.aws.json to config/ section
19
+ - Removed CLAUDE.md and AGENTS.md from package root (only templates exist)
20
+ - Added speckit/ directory to structure
21
+
22
+ - **config/AGENTS.md:**
23
+ - Updated all references: `.claude/docs/` → `.claude/config/`
24
+ - Fixed quick links and support documentation paths
25
+
26
+ - **config/agent-catalog.md:**
27
+ - Updated all 5 context contract references: `.claude/docs/` → `.claude/config/`
28
+
29
+ - **index.js:**
30
+ - Deprecated `getDocPath()` function with console warning
31
+ - Function now redirects to `config/` directory instead of non-existent `docs/`
32
+ - Added JSDoc @deprecated annotation
33
+
34
+ - **README.en.md (Documentation section):**
35
+ - Removed broken reference to `./CLAUDE.md` (file not in package)
36
+ - Fixed all documentation links: `./docs/` → `./config/`
37
+ - Updated to match actual config/ directory structure
38
+
39
+ - **speckit/README.en.md:**
40
+ - Removed 3 non-existent commands: speckit.clarify, speckit.analyze-plan, speckit.constitution
41
+ - Updated command count: 9 → 7 actual commands
42
+ - Removed references to non-existent tasks-richer.py tool
43
+ - Removed entire sections for non-existent templates (data-model-template.md, contracts-template.md)
44
+ - Updated tool files list with actual tools (task_manager.py, clarify_engine.py, context_provider.py)
45
+ - Fixed all code examples to use only existing commands
46
+
47
+ - **tools/context_provider.py:**
48
+ - Added auto-detection for project-context.json location
49
+ - Honors GAIA_CONTEXT_PATH environment variable
50
+ - Falls back through common locations (.claude/project-context.json, .claude/project-context/project-context.json)
51
+ - Fixes agent routing failures when project-context.json is in non-legacy location
52
+
53
+ - **package.json:**
54
+ - Fixed `npm test` script (was calling non-existent pytest tests)
55
+ - Now echoes informative message about fixture availability
56
+
57
+ - **Agent Branding Unification:**
58
+ - Renamed `agents/claude-architect.md` → `agents/gaia.md` (aligns with gaia-ops package name)
59
+ - Renamed `commands/gaina.md` → `commands/gaia.md` (unified as `/gaia` command)
60
+ - Updated all references in README.md, README.en.md, and agents/gaia.md
61
+ - Complete branding consistency: package name, agent name, and command name all use "gaia"
62
+
63
+ ### Benefits
64
+ - ✅ **Accurate documentation:** All paths and structures match actual package contents
65
+ - ✅ **No broken links:** References point to existing files
66
+ - ✅ **Clear API:** Deprecated functions clearly marked
67
+ - ✅ **User trust:** Documentation matches reality
68
+ - ✅ **npm test passes:** No false failures
69
+
70
+ ---
71
+
10
72
  ## [2.2.0] - 2025-11-10
11
73
 
12
74
  ### Added - Unified Settings Template & Auto-Installation
@@ -46,7 +108,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
46
108
  - Updated header and JSDoc comments with new package name
47
109
  - Updated example usage
48
110
 
49
- - **agents/claude-architect.md:**
111
+ - **agents/gaia.md:**
50
112
  - Updated system paths to reflect gaia-ops package structure
51
113
  - Clarified symlink architecture and layout
52
114
 
package/README.en.md CHANGED
@@ -15,8 +15,8 @@ Multi-agent orchestration system for Claude Code - DevOps automation toolkit.
15
15
  ### Features
16
16
 
17
17
  - **Multi-cloud support** - Works with GCP, AWS, and ready for Azure
18
- - **6 specialist agents** (terraform-architect, gitops-operator, gcp-troubleshooter, aws-troubleshooter, devops-developer, claude-architect)
19
- - **3 meta-agents** (Explore, Plan, claude-architect)
18
+ - **6 specialist agents** (terraform-architect, gitops-operator, gcp-troubleshooter, aws-troubleshooter, devops-developer, Gaia)
19
+ - **3 meta-agents** (Explore, Plan, Gaia)
20
20
  - **Clarification engine** for ambiguity detection
21
21
  - **Approval gates** for T3 operations (terraform apply, kubectl apply, etc.)
22
22
  - **Git commit validation** with Conventional Commits
@@ -91,7 +91,7 @@ node_modules/@jaguilar87/gaia-ops/
91
91
  │ ├── gcp-troubleshooter.md
92
92
  │ ├── aws-troubleshooter.md
93
93
  │ ├── devops-developer.md
94
- │ └── claude-architect.md
94
+ │ └── gaia.md
95
95
  ├── tools/ # Orchestration tools
96
96
  │ ├── context_provider.py
97
97
  │ ├── agent_router.py
@@ -99,28 +99,35 @@ node_modules/@jaguilar87/gaia-ops/
99
99
  │ ├── approval_gate.py
100
100
  │ ├── commit_validator.py
101
101
  │ └── task_manager.py
102
- ├── hooks/ # Git hooks
103
- └── pre-commit
102
+ ├── hooks/ # Claude Code hooks (Python)
103
+ ├── pre_tool_use.py
104
+ │ ├── post_tool_use.py
105
+ │ ├── subagent_stop.py
106
+ │ ├── session_start.py
107
+ │ └── pre_kubectl_security.py
104
108
  ├── commands/ # Slash commands
105
- │ ├── architect.md
106
- └── speckit.*.md
109
+ │ ├── gaia.md
110
+ ├── save-session.md
111
+ │ ├── restore-session.md
112
+ │ ├── session-status.md
113
+ │ └── speckit.*.md (7 commands)
107
114
  ├── config/ # Configuration & documentation
108
115
  │ ├── AGENTS.md
109
116
  │ ├── orchestration-workflow.md
110
117
  │ ├── git-standards.md
111
118
  │ ├── context-contracts.md
119
+ │ ├── context-contracts.gcp.json
120
+ │ ├── context-contracts.aws.json
112
121
  │ ├── agent-catalog.md
113
122
  │ └── git_standards.json
114
- ├── templates/ # Code templates
123
+ ├── templates/ # Installation templates
115
124
  │ ├── CLAUDE.template.md
116
- │ └── code-examples/
117
- ├── commit_validation.py
118
- ├── clarification_workflow.py
119
- └── approval_gate_workflow.py
120
- ├── config/ # Configuration
121
- │ └── git_standards.json
122
- ├── CLAUDE.md # Core orchestrator instructions
123
- ├── AGENTS.md # System overview
125
+ │ └── settings.template.json
126
+ ├── speckit/ # Spec-Kit methodology
127
+ ├── README.md
128
+ ├── templates/
129
+ ├── scripts/
130
+ │ └── governance.md
124
131
  ├── CHANGELOG.md # Version history
125
132
  ├── package.json
126
133
  └── index.js # Helper functions
@@ -182,18 +189,17 @@ This package follows [Semantic Versioning](https://semver.org/):
182
189
  - **MINOR:** New features, agents, or improvements
183
190
  - **PATCH:** Bug fixes, clarifications, typos
184
191
 
185
- Current version: **2.1.0**
192
+ Current version: **2.2.0**
186
193
 
187
194
  See [CHANGELOG.md](./CHANGELOG.md) for version history.
188
195
 
189
196
  ## Documentation
190
197
 
191
- - **Core Instructions:** [CLAUDE.md](./CLAUDE.md) (154 lines)
192
- - **System Overview:** [AGENTS.md](./AGENTS.md) (95 lines)
193
- - **Orchestration Workflow:** [docs/orchestration-workflow.md](./docs/orchestration-workflow.md) (735 lines)
194
- - **Git Standards:** [docs/git-standards.md](./docs/git-standards.md) (682 lines)
195
- - **Context Contracts:** [docs/context-contracts.md](./docs/context-contracts.md) (673 lines)
196
- - **Agent Catalog:** [docs/agent-catalog.md](./docs/agent-catalog.md) (603 lines)
198
+ - **System Overview:** [config/AGENTS.md](./config/AGENTS.md) (95 lines)
199
+ - **Orchestration Workflow:** [config/orchestration-workflow.md](./config/orchestration-workflow.md) (735 lines)
200
+ - **Git Standards:** [config/git-standards.md](./config/git-standards.md) (682 lines)
201
+ - **Context Contracts:** [config/context-contracts.md](./config/context-contracts.md) (673 lines)
202
+ - **Agent Catalog:** [config/agent-catalog.md](./config/agent-catalog.md) (603 lines)
197
203
 
198
204
  ## Requirements
199
205
 
package/README.md CHANGED
@@ -15,8 +15,8 @@ Sistema de orquestación multi-agente para Claude Code - Toolkit de automatizaci
15
15
  ### Características
16
16
 
17
17
  - **Soporte multi-cloud** - Funciona con GCP, AWS, y listo para Azure
18
- - **6 agentes especialistas** (terraform-architect, gitops-operator, gcp-troubleshooter, aws-troubleshooter, devops-developer, claude-architect)
19
- - **3 meta-agentes** (Explore, Plan, claude-architect)
18
+ - **6 agentes especialistas** (terraform-architect, gitops-operator, gcp-troubleshooter, aws-troubleshooter, devops-developer, Gaia)
19
+ - **3 meta-agentes** (Explore, Plan, Gaia)
20
20
  - **Motor de clarificación** para detección de ambigüedades
21
21
  - **Puertas de aprobación** para operaciones T3 (terraform apply, kubectl apply, etc.)
22
22
  - **Validación de commits Git** con Conventional Commits
@@ -91,7 +91,7 @@ node_modules/@jaguilar87/gaia-ops/
91
91
  │ ├── gcp-troubleshooter.md
92
92
  │ ├── aws-troubleshooter.md
93
93
  │ ├── devops-developer.md
94
- │ └── claude-architect.md
94
+ │ └── gaia.md
95
95
  ├── tools/ # Herramientas de orquestación
96
96
  │ ├── context_provider.py
97
97
  │ ├── agent_router.py
@@ -99,28 +99,35 @@ node_modules/@jaguilar87/gaia-ops/
99
99
  │ ├── approval_gate.py
100
100
  │ ├── commit_validator.py
101
101
  │ └── task_manager.py
102
- ├── hooks/ # Git hooks
103
- └── pre-commit
102
+ ├── hooks/ # Hooks de Claude Code (Python)
103
+ ├── pre_tool_use.py
104
+ │ ├── post_tool_use.py
105
+ │ ├── subagent_stop.py
106
+ │ ├── session_start.py
107
+ │ └── pre_kubectl_security.py
104
108
  ├── commands/ # Comandos slash
105
- │ ├── architect.md
106
- └── speckit.*.md
109
+ │ ├── gaia.md
110
+ ├── save-session.md
111
+ │ ├── restore-session.md
112
+ │ ├── session-status.md
113
+ │ └── speckit.*.md (7 comandos)
107
114
  ├── config/ # Configuración y documentación
108
115
  │ ├── AGENTS.md
109
116
  │ ├── orchestration-workflow.md
110
117
  │ ├── git-standards.md
111
118
  │ ├── context-contracts.md
119
+ │ ├── context-contracts.gcp.json
120
+ │ ├── context-contracts.aws.json
112
121
  │ ├── agent-catalog.md
113
122
  │ └── git_standards.json
114
- ├── templates/ # Plantillas de código
123
+ ├── templates/ # Templates de instalación
115
124
  │ ├── CLAUDE.template.md
116
- │ └── code-examples/
117
- ├── commit_validation.py
118
- ├── clarification_workflow.py
119
- └── approval_gate_workflow.py
120
- ├── config/ # Configuración
121
- │ └── git_standards.json
122
- ├── CLAUDE.md # Instrucciones del orquestador principal
123
- ├── AGENTS.md # Vista general del sistema
125
+ │ └── settings.template.json
126
+ ├── speckit/ # Metodología Spec-Kit
127
+ ├── README.md
128
+ ├── templates/
129
+ ├── scripts/
130
+ │ └── governance.md
124
131
  ├── CHANGELOG.md # Historial de versiones
125
132
  ├── package.json
126
133
  └── index.js # Funciones auxiliares
@@ -182,7 +189,7 @@ Este paquete sigue [Versionamiento Semántico](https://semver.org/):
182
189
  - **MINOR:** Nuevas características, agentes o mejoras
183
190
  - **PATCH:** Correcciones de bugs, clarificaciones, errores tipográficos
184
191
 
185
- Versión actual: **2.1.0**
192
+ Versión actual: **2.2.0**
186
193
 
187
194
  Ver [CHANGELOG.md](./CHANGELOG.md) para el historial de versiones.
188
195
 
@@ -1,11 +1,11 @@
1
1
  ---
2
- name: claude-architect
3
- description: A meta-agent specialized in analyzing, diagnosing, and optimizing the intelligent agent orchestration system itself. It understands the system architecture, analyzes logs/metrics, researches best practices, and proposes improvements.
2
+ name: gaia
3
+ description: Gaia is the meta-agent for the gaia-ops ecosystem—its purpose is to understand, document, and continuously optimize every component of the orchestrator, agents, commands, templates, sample stacks, and documentation that ship in this package and the consuming projects that symlink to it.
4
4
  tools: Read, Glob, Grep, Bash, Task, WebSearch, Python
5
5
  model: inherit
6
6
  ---
7
7
 
8
- You are a senior system architect and AI agent systems specialist. Your unique purpose is to **analyze and optimize the intelligent agent orchestration system itself** - acting as a meta-layer that understands how the orchestrator, agents, router, context provider, and all system components work together.
8
+ You are Gaia, the senior system architect and AI agent systems specialist for the gaia-ops stack. Your unique purpose is to **analyze and optimize Gaia’s intelligent agent orchestration system end-to-end**—acting as a meta-layer that understands how the orchestrator, agents, commands, router, context provider, Spec-Kit assets, and all system components work together across repositories (package root, downstream project `.claude/`, ops symlinks).
9
9
 
10
10
  ## ⚡ QUICK START - Read This First
11
11
 
@@ -18,10 +18,10 @@ You are a senior system architect and AI agent systems specialist. Your unique p
18
18
  **Where Everything Lives (Package + Symlink Layout):**
19
19
  - 🏗️ Package root: `/home/jaguilar/aaxis/rnd/repositories/gaia-ops/` → mirrors `node_modules/@jaguilar87/gaia-ops/` when installed
20
20
  - 📋 Orchestrator: `CLAUDE.md` at the package root (templated into consuming repos)
21
- - 🤖 Agents: `agents/*.md` (6 specialists: terraform-architect, gitops-operator, gcp-troubleshooter, aws-troubleshooter, devops-developer, claude-architect)
21
+ - 🤖 Agents: `agents/*.md` (6 specialists: terraform-architect, gitops-operator, gcp-troubleshooter, aws-troubleshooter, devops-developer, gaia)
22
22
  - 🛠️ Tools: `tools/` (context_provider.py, agent_router.py, clarify_engine.py, approval_gate.py, commit_validator.py, task_manager.py)
23
23
  - 📚 Config docs: `config/` (AGENTS, orchestration-workflow, git-standards, context-contracts, agent-catalog)
24
- - 🗂️ Commands: `commands/*.md` (`/architect`, `/save-session`, `/session-status`, `/speckit.*`)
24
+ - 🗂️ Commands: `commands/*.md` (`/gaia`, `/save-session`, `/session-status`, `/speckit.*`)
25
25
  - 🎯 Spec-Kit assets: `speckit/README*.md`, `speckit/templates/`, `speckit/scripts/`, `speckit/governance.md`, `speckit/decisions/`
26
26
  - 🔧 Reference stacks: `terraform/`, `gitops/`, `app-services/` illustrate how agents interact with user IaC/App repos
27
27
  - 💾 Project data: consuming repos host `.claude/project-context.json`, `.claude/logs/`, `.claude/tests/`, while `ops/` carries shared symlink helpers
@@ -73,7 +73,7 @@ gaia-ops/ (mirrors node_modules/@jaguilar87/gaia-ops/ and symlinks into project
73
73
  │ ├── gcp-troubleshooter.md
74
74
  │ ├── aws-troubleshooter.md
75
75
  │ ├── devops-developer.md
76
- │ └── claude-architect.md
76
+ │ └── gaia.md (this file)
77
77
  ├── tools/ # System intelligence + automation
78
78
  │ ├── context_provider.py # Deterministic context generation
79
79
  │ ├── agent_router.py # Semantic routing (92.7% target accuracy)
@@ -1,13 +1,13 @@
1
1
  ---
2
- description: Invoke claude-architect to analyze system architecture, diagnose issues, or propose improvements
2
+ description: Invoke Gaia to analyze system architecture, diagnose issues, or propose improvements
3
3
  scope: project
4
4
  ---
5
5
 
6
- You are invoking the **claude-architect** meta-agent to analyze the agent orchestration system itself.
6
+ You are invoking **Gaia**, the gaia-ops meta-agent that analyzes the agent orchestration system itself.
7
7
 
8
8
  ## Context: System Expert Agent
9
9
 
10
- claude-architect is a specialized agent that understands:
10
+ Gaia is a specialized agent that understands:
11
11
  - The entire agent system architecture (.claude/, CLAUDE.md, agents, tools, hooks)
12
12
  - Spec-Kit workflows (specify, plan, tasks, implement, add-task, analyze-task)
13
13
  - Session management (save, restore, bundles)
@@ -18,7 +18,7 @@ claude-architect is a specialized agent that understands:
18
18
 
19
19
  ## Your Task
20
20
 
21
- Invoke the claude-architect agent with the user's request. The agent will proactively read system files, analyze logs, research best practices, and provide comprehensive analysis with actionable recommendations.
21
+ Invoke Gaia with the user's request. The agent will proactively read system files, analyze logs, research best practices, and provide comprehensive analysis with actionable recommendations tied to gaia-ops.
22
22
 
23
23
  **IMPORTANT:** Pass the user's exact question/request to the agent. Examples:
24
24
  - "Analiza este log y dime qué pasó"
@@ -29,11 +29,11 @@ Invoke the claude-architect agent with the user's request. The agent will proact
29
29
 
30
30
  ## Invocation
31
31
 
32
- Use the Task tool to invoke claude-architect with this structure:
32
+ Use the Task tool to invoke Gaia with this structure:
33
33
 
34
34
  ```
35
35
  Task(
36
- subagent_type="claude-architect",
36
+ subagent_type="gaia",
37
37
  description="[Brief 3-5 word description]",
38
38
  prompt="""
39
39
  ## System Context (Auto-provided)
package/config/AGENTS.md CHANGED
@@ -8,9 +8,9 @@ See `CLAUDE.md` for complete orchestrator instructions.
8
8
 
9
9
  **Quick links:**
10
10
  - Core principles: `CLAUDE.md` (lines 18-36)
11
- - Workflow: `.claude/docs/orchestration-workflow.md`
12
- - Git standards: `.claude/docs/git-standards.md`
13
- - Agent catalog: `.claude/docs/agent-catalog.md`
11
+ - Workflow: `.claude/config/orchestration-workflow.md`
12
+ - Git standards: `.claude/config/git-standards.md`
13
+ - Agent catalog: `.claude/config/agent-catalog.md`
14
14
 
15
15
  ---
16
16
 
@@ -71,7 +71,7 @@ Claude Code (Orchestrator)
71
71
  ├── gcp-troubleshooter (GCP diagnostics)
72
72
  ├── aws-troubleshooter (AWS diagnostics)
73
73
  ├── devops-developer (Application build/test)
74
- ├── claude-architect (System optimization)
74
+ ├── Gaia (System optimization)
75
75
  ├── Explore (Codebase exploration)
76
76
  └── Plan (Implementation planning)
77
77
  ```
@@ -153,7 +153,7 @@ See `CLAUDE.md` and `.claude/CHANGELOG.md` for contribution guidelines.
153
153
 
154
154
  - **Issues:** Create issue in GitHub repository
155
155
  - **Questions:** Contact Jorge Aguilar (jaguilar@aaxis.com)
156
- - **Documentation:** See `.claude/docs/*.md`
156
+ - **Documentation:** See `.claude/config/*.md`
157
157
 
158
158
  ---
159
159
 
@@ -30,7 +30,7 @@ These agents work on **CLAUDE CODE SYSTEM ITSELF** (agent orchestration, tooling
30
30
 
31
31
  | Agent | Primary Role | Context Type | Model |
32
32
  |-------|--------------|--------------|-------|
33
- | **claude-architect** | System analysis & optimization | Manual (logs, .claude/, tests) | inherit |
33
+ | **Gaia** | System analysis & optimization | Manual (gaia-ops package, .claude/, tests) | inherit |
34
34
  | **Explore** | Codebase exploration for understanding | Automatic (file patterns) | haiku |
35
35
  | **Plan** | Planning mode for implementation design | Automatic (user prompt) | inherit |
36
36
 
@@ -82,7 +82,7 @@ These agents work on **CLAUDE CODE SYSTEM ITSELF** (agent orchestration, tooling
82
82
  - T2: `terraform plan`
83
83
  - T3: `terragrunt apply` (requires approval)
84
84
 
85
- **Context Contract:** See `.claude/docs/context-contracts.md#terraform-architect`
85
+ **Context Contract:** See `.claude/config/context-contracts.md#terraform-architect`
86
86
 
87
87
  **Use Cases:**
88
88
  - "Create a CloudSQL instance with Terraform"
@@ -148,7 +148,7 @@ Task(
148
148
  - T2: `git push` to feature branch, `flux reconcile`
149
149
  - T3: `kubectl apply`, `git push` to main (requires approval)
150
150
 
151
- **Context Contract:** See `.claude/docs/context-contracts.md#gitops-operator`
151
+ **Context Contract:** See `.claude/config/context-contracts.md#gitops-operator`
152
152
 
153
153
  **Use Cases:**
154
154
  - "Deploy tcm-api service to non-prod cluster"
@@ -214,7 +214,7 @@ Task(
214
214
  - T2: `gcloud compute ssh` (for VM diagnostics)
215
215
  - T3: None (diagnostic agent, no destructive operations)
216
216
 
217
- **Context Contract:** See `.claude/docs/context-contracts.md#gcp-troubleshooter`
217
+ **Context Contract:** See `.claude/config/context-contracts.md#gcp-troubleshooter`
218
218
 
219
219
  **Use Cases:**
220
220
  - "Why is tcm-api pod crashing with database connection error?"
@@ -270,7 +270,7 @@ Task(
270
270
  - T2: `aws ec2 ssh` (for EC2 diagnostics)
271
271
  - T3: None (diagnostic agent, no destructive operations)
272
272
 
273
- **Context Contract:** See `.claude/docs/context-contracts.md#aws-troubleshooter`
273
+ **Context Contract:** See `.claude/config/context-contracts.md#aws-troubleshooter`
274
274
 
275
275
  **Use Cases:**
276
276
  - "Why is EKS pod failing with IAM permission error?"
@@ -308,7 +308,7 @@ Task(
308
308
  - T2: `git push` to feature branch, `npm publish` (to test registry)
309
309
  - T3: `git push` to main (requires approval)
310
310
 
311
- **Context Contract:** See `.claude/docs/context-contracts.md#devops-developer`
311
+ **Context Contract:** See `.claude/config/context-contracts.md#devops-developer`
312
312
 
313
313
  **Use Cases:**
314
314
  - "Run tests and fix failures"
@@ -331,7 +331,7 @@ Task(
331
331
 
332
332
  ## Meta-Agents: Detailed Catalog
333
333
 
334
- ### claude-architect
334
+ ### Gaia
335
335
 
336
336
  **Full Name:** Claude Code System Architect
337
337
 
@@ -363,7 +363,7 @@ Task(
363
363
  ```python
364
364
  # Orchestrator provides manual context in prompt
365
365
  Task(
366
- subagent_type="claude-architect",
366
+ subagent_type="gaia",
367
367
  description="Analyze routing accuracy",
368
368
  prompt="""
369
369
  ## System Context
@@ -471,7 +471,7 @@ User Request
471
471
  │ └─ YES → devops-developer
472
472
 
473
473
  ├─ System/agent analysis?
474
- │ └─ YES → claude-architect
474
+ │ └─ YES → Gaia
475
475
 
476
476
  ├─ Codebase exploration/understanding?
477
477
  │ └─ YES → Explore
@@ -488,8 +488,8 @@ User Request
488
488
 
489
489
  ```python
490
490
  # WRONG
491
- context = context_provider.py("claude-architect", "analyze logs")
492
- Task(subagent_type="claude-architect", prompt=context)
491
+ context = context_provider.py("gaia", "analyze logs")
492
+ Task(subagent_type="gaia", prompt=context)
493
493
  ```
494
494
 
495
495
  **Why:** Meta-agents work on the SYSTEM, not projects. They need system paths, not project context.
@@ -497,7 +497,7 @@ Task(subagent_type="claude-architect", prompt=context)
497
497
  **Correct:**
498
498
  ```python
499
499
  Task(
500
- subagent_type="claude-architect",
500
+ subagent_type="gaia",
501
501
  prompt=f"System path: {system_path}\n\nTask: analyze logs"
502
502
  )
503
503
  ```
@@ -576,7 +576,7 @@ Track in `.claude/logs/agent-metrics.jsonl`:
576
576
  | gcp-troubleshooter | 123 | 98% | 28s | N/A |
577
577
  | aws-troubleshooter | 45 | 96% | 31s | N/A |
578
578
  | devops-developer | 189 | 92% | 18s | 15% |
579
- | claude-architect | 34 | 100% | 120s | N/A |
579
+ | Gaia | 34 | 100% | 120s | N/A |
580
580
  | Explore | 456 | 99% | 8s | N/A |
581
581
  | Plan | 78 | 95% | 22s | N/A |
582
582
 
@@ -601,4 +601,4 @@ Track in `.claude/logs/agent-metrics.jsonl`:
601
601
 
602
602
  ### 1.x (Historical)
603
603
  - Embedded in CLAUDE.md
604
- - Basic agent list with roles
604
+ - Basic agent list with roles
@@ -526,7 +526,7 @@ devops-developer:
526
526
 
527
527
  ---
528
528
 
529
- ## claude-architect (Meta-Agent)
529
+ ## Gaia (Meta-Agent)
530
530
 
531
531
  **Purpose:** Analyzes, diagnoses, and optimizes the agent orchestration system itself
532
532
 
@@ -535,7 +535,7 @@ devops-developer:
535
535
  ### Manual Context Structure
536
536
 
537
537
  ```yaml
538
- claude-architect:
538
+ gaia:
539
539
  context_type: "manual"
540
540
 
541
541
  provided_in_prompt:
@@ -561,7 +561,7 @@ claude-architect:
561
561
 
562
562
  ```python
563
563
  Task(
564
- subagent_type="claude-architect",
564
+ subagent_type="gaia",
565
565
  description="Analyze routing accuracy",
566
566
  prompt="""
567
567
  ## System Context
@@ -714,7 +714,7 @@ if agent_contract_version == "1.0":
714
714
  | gcp-troubleshooter | project_details, terraform_infrastructure, gitops_configuration | application_services, issue_context | GCP diagnostics |
715
715
  | aws-troubleshooter | project_details, terraform_infrastructure, gitops_configuration | application_services, issue_context | AWS diagnostics |
716
716
  | devops-developer | project_details, operational_guidelines | application_services, development_context | App build/test/debug |
717
- | claude-architect | (manual context in prompt) | N/A | System analysis & optimization |
717
+ | Gaia | (manual context in prompt) | N/A | System analysis & optimization |
718
718
 
719
719
  ---
720
720
 
package/index.js CHANGED
@@ -56,11 +56,13 @@ export function getCommandPath(commandName) {
56
56
 
57
57
  /**
58
58
  * Get absolute path to documentation
59
+ * @deprecated Use getConfigPath() instead. Documentation moved from docs/ to config/ in v2.0.0
59
60
  * @param {string} docName - Name of the doc (e.g., 'orchestration-workflow.md')
60
61
  * @returns {string} Absolute path to doc file
61
62
  */
62
63
  export function getDocPath(docName) {
63
- return join(PACKAGE_ROOT, 'docs', docName);
64
+ console.warn('getDocPath() is deprecated. Use getConfigPath() instead. Documentation is now in config/ directory.');
65
+ return join(PACKAGE_ROOT, 'config', docName);
64
66
  }
65
67
 
66
68
  /**
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@jaguilar87/gaia-ops",
3
- "version": "2.2.0",
3
+ "version": "2.2.1",
4
4
  "description": "Multi-agent orchestration system for Claude Code - DevOps automation toolkit",
5
5
  "main": "index.js",
6
6
  "type": "module",
@@ -44,7 +44,7 @@
44
44
  "index.js"
45
45
  ],
46
46
  "scripts": {
47
- "test": "pytest tests/ -v",
47
+ "test": "echo 'No tests configured. Fixtures available in tests/fixtures/'",
48
48
  "validate": "python3 tools/commit_validator.py",
49
49
  "lint": "eslint .",
50
50
  "clean": "find . -type d -name '__pycache__' -exec rm -rf {} + 2>/dev/null || true",
@@ -60,26 +60,23 @@ Spec-Kit provides structured workflow for feature development:
60
60
  ├── templates/ # 5 markdown templates
61
61
  │ ├── spec-template.md # Feature specification template
62
62
  │ ├── plan-template.md # Implementation plan template
63
- │ ├── tasks-template.md # Tasks list template
64
- │ ├── data-model-template.md # Data model template
65
- │ └── contracts-template.md # API contracts template
66
- └── memory/ # Legacy directory (deprecated)
67
- └── constitution.md # MOVED to project root
63
+ │ ├── tasks-template.md # Task list template
64
+ │ ├── adr-template.md # Architecture Decision Record template
65
+ │ └── agent-file-template.md # Agent context file template
68
66
 
69
- .claude/commands/ # 9 /speckit.* commands
67
+ .claude/commands/ # 7 /speckit.* commands
70
68
  ├── speckit.specify.md # Create specification
71
- ├── speckit.clarify.md # Clarify ambiguities
69
+ ├── speckit.init.md # Initialize Spec-Kit structure
72
70
  ├── speckit.plan.md # Create implementation plan
73
71
  ├── speckit.tasks.md # Generate task list
74
- ├── speckit.analyze-plan.md # Validate consistency (cross-artifact)
75
72
  ├── speckit.analyze-task.md # Analyze specific task (deep-dive)
76
73
  ├── speckit.implement.md # Execute implementation
77
- ├── speckit.add-task.md # Add ad-hoc task (with auto-validation)
78
- └── speckit.constitution.md # Update constitution
74
+ └── speckit.add-task.md # Add ad-hoc task (with auto-validation)
79
75
 
80
76
  .claude/tools/ # Python utilities
81
77
  ├── agent_router.py # Route tasks to agents
82
- └── tasks-richer.py # Auto-enrich tasks with metadata
78
+ ├── task_manager.py # Manage task lifecycle
79
+ └── clarify_engine.py # Ambiguity detection
83
80
 
84
81
  <project-root>/ # User-specified root (e.g., spec-kit-tcm-plan/)
85
82
  ├── constitution.md # Project governance principles
@@ -113,9 +110,9 @@ Spec-Kit provides structured workflow for feature development:
113
110
  mkdir -p spec-kit-tcm-plan/specs
114
111
  ```
115
112
 
116
- **Step 2: Create constitution (optional)**
113
+ **Step 2: Initialize Spec-Kit structure**
117
114
  ```bash
118
- /speckit.constitution spec-kit-tcm-plan
115
+ /speckit.init spec-kit-tcm-plan
119
116
  ```
120
117
 
121
118
  **Ready!** Commands are available immediately. Example:
@@ -133,29 +130,24 @@ mkdir -p spec-kit-tcm-plan/specs
133
130
 
134
131
  | Command | Syntax | Purpose | When to Use |
135
132
  |---------|--------|---------|-------------|
133
+ | **init** | `/speckit.init <root>` | Bootstrap Spec-Kit structure | Initial project setup |
136
134
  | **specify** | `/speckit.specify <root> "description"` | Create new feature specification | Start of workflow |
137
- | **clarify** | `/speckit.clarify <root> <feature>` | Resolve ambiguities in spec.md | After specify, before plan (optional) |
138
- | **plan** | `/speckit.plan <root> <feature>` | Create technical implementation plan | After specify/clarify |
135
+ | **plan** | `/speckit.plan <root> <feature>` | Create technical implementation plan | After specify |
139
136
  | **tasks** | `/speckit.tasks <root> <feature>` | Generate task list with metadata | After plan |
140
- | **analyze-plan** | `/speckit.analyze-plan <root> <feature>` | Validate spec/plan/tasks consistency | After tasks, before implement (optional) |
141
137
  | **implement** | `/speckit.implement <root> <feature>` | Execute tasks with automatic routing | After tasks |
142
138
  | **add-task** | `/speckit.add-task <root> <feature> "desc"` | Add ad-hoc task with validation | During implement |
143
139
  | **analyze-task** | `/speckit.analyze-task <root> <feature> T###` | Deep analysis of specific task | Before executing risky tasks |
144
- | **constitution** | `/speckit.constitution <root>` | Create/update governance principles | Initial setup or updates |
145
140
 
146
141
  ### Usage Examples
147
142
 
148
143
  ```bash
149
144
  # Basic complete workflow
145
+ /speckit.init spec-kit-tcm-plan
150
146
  /speckit.specify spec-kit-tcm-plan "Project Guidance Deployment"
151
147
  /speckit.plan spec-kit-tcm-plan 004-project-guidance-deployment
152
148
  /speckit.tasks spec-kit-tcm-plan 004-project-guidance-deployment
153
149
  /speckit.implement spec-kit-tcm-plan 004-project-guidance-deployment
154
150
 
155
- # With optional validation
156
- /speckit.clarify spec-kit-tcm-plan 004-project-guidance-deployment
157
- /speckit.analyze-plan spec-kit-tcm-plan 004-project-guidance-deployment
158
-
159
151
  # During implementation
160
152
  /speckit.add-task spec-kit-tcm-plan 004-project-guidance-deployment "Fix config error"
161
153
  /speckit.analyze-task spec-kit-tcm-plan 004-project-guidance-deployment T042
@@ -181,43 +173,12 @@ Location: `.claude/speckit/scripts/`
181
173
  **Format:**
182
174
  ```markdown
183
175
  - [ ] T001 Task description
184
- <!-- Metadata injected by tasks-richer.py -->
185
176
  ```
186
177
 
187
178
  **Used by:** `/speckit.tasks`
188
179
 
189
180
  ---
190
181
 
191
- ### data-model-template.md
192
-
193
- **Purpose:** Data model documentation template
194
-
195
- **Location:** `.claude/speckit/templates/data-model-template.md`
196
-
197
- **Sections:**
198
- - Entity Definitions
199
- - Relationships
200
- - Schema Design
201
- - Migrations
202
-
203
- **Optional:** Created manually when needed
204
-
205
- ---
206
-
207
- ### contracts-template.md
208
-
209
- **Purpose:** API contracts template
210
-
211
- **Location:** `.claude/speckit/templates/contracts-template.md`
212
-
213
- **Sections:**
214
- - API Endpoints
215
- - Request/Response Schemas
216
- - Error Codes
217
- - Authentication
218
-
219
- **Optional:** Created manually when needed
220
-
221
182
  ## Auto-Enrichment
222
183
 
223
184
  ### What is Auto-Enrichment?
@@ -234,12 +195,7 @@ Automatic injection of metadata into tasks for agent routing and risk assessment
234
195
 
235
196
  ### Enrichment Process
236
197
 
237
- **Step 1: Task parsing**
238
- ```bash
239
- python3 .claude/tools/tasks-richer.py tasks.md
240
- ```
241
-
242
- **Step 2: Agent routing**
198
+ **Step 1: Agent routing**
243
199
  ```bash
244
200
  python3 .claude/tools/agent_router.py --json "Task description"
245
201
  ```
@@ -411,10 +367,7 @@ WARNING: constitution.md not found at spec-kit-tcm-plan/constitution.md
411
367
  **Solution:**
412
368
  ```bash
413
369
  # Create constitution
414
- /speckit.constitution
415
-
416
- # Or move existing
417
- mv .claude/speckit/memory/constitution.md spec-kit-tcm-plan/
370
+ # Create governance document manually if needed
418
371
  ```
419
372
 
420
373
  ---
@@ -561,8 +514,8 @@ jq --version
561
514
  ### Feature Development
562
515
 
563
516
  - ✅ Follow workflow order (specify → plan → tasks → implement)
564
- - ✅ Use `/speckit.clarify` to resolve ambiguities early
565
- - ✅ Run `/speckit.analyze` before implementation (optional but recommended)
517
+ - ✅ Use clarify_engine.py for ambiguity detection (automatic)
518
+ - ✅ Run `/speckit.analyze-task` for high-risk tasks before execution
566
519
  - ✅ Let auto-enrichment handle metadata (don't edit manually)
567
520
 
568
521
  ### Risk Management
@@ -600,20 +553,18 @@ jq --version
600
553
  All commands in `.claude/commands/speckit.*.md`:
601
554
  - speckit.init.md
602
555
  - speckit.specify.md
603
- - speckit.clarify.md
604
556
  - speckit.plan.md
605
557
  - speckit.tasks.md
606
- - speckit.analyze-plan.md
607
558
  - speckit.analyze-task.md
608
559
  - speckit.implement.md
609
560
  - speckit.add-task.md
610
- - speckit.constitution.md
611
561
 
612
562
  ### Tool Files
613
563
 
614
564
  - `.claude/tools/agent_router.py` - Agent routing logic
615
- - `.claude/tools/tasks-richer.py` - Task enrichment logic
616
- - `.claude/tools/context_section_reader.py` - Context filtering
565
+ - `.claude/tools/task_manager.py` - Task lifecycle management
566
+ - `.claude/tools/clarify_engine.py` - Ambiguity detection
567
+ - `.claude/tools/context_provider.py` - Context provisioning
617
568
 
618
569
  **Framework Base**
619
570
 
@@ -86,7 +86,7 @@ Guidance for Claude Code orchestrator working in this repository.
86
86
  | gitops-operator | project_details, gitops_configuration, cluster_details, operational_guidelines |
87
87
  | gcp/aws-troubleshooter | project_details, terraform_infrastructure, gitops_configuration, application_services |
88
88
  | devops-developer | project_details, operational_guidelines |
89
- | claude-architect | Manual context (system paths, logs, tests) |
89
+ | Gaia | Manual context (gaia-ops paths, logs, tests) |
90
90
 
91
91
  ## Agent System
92
92
 
@@ -106,7 +106,7 @@ Guidance for Claude Code orchestrator working in this repository.
106
106
 
107
107
  | Agent | Primary Role |
108
108
  |-------|--------------|
109
- | **claude-architect** | System analysis & optimization |
109
+ | **Gaia** | System analysis & optimization |
110
110
  | **Explore** | Codebase exploration |
111
111
  | **Plan** | Implementation planning |
112
112
 
@@ -6,9 +6,30 @@ from pathlib import Path
6
6
  from typing import Dict, List, Any, Optional
7
7
 
8
8
  # This script is expected to be run from the root of the repository.
9
- # The project context file is located at `.claude/project-context.json`.
10
- # We construct the path relative to the script's assumed execution location.
11
- DEFAULT_CONTEXT_PATH = Path(".claude/project-context.json")
9
+ # Historically the project context lived at `.claude/project-context.json`, but
10
+ # newer installs keep it under `.claude/project-context/project-context.json`.
11
+ # We auto-detect the most common locations (and honor GAIA_CONTEXT_PATH) so agent
12
+ # routing never breaks if the file isn't symlinked to the legacy path.
13
+ def get_default_context_path() -> Path:
14
+ """Detects the project-context.json location, honoring GAIA_CONTEXT_PATH."""
15
+ env_override = os.environ.get("GAIA_CONTEXT_PATH")
16
+ if env_override:
17
+ return Path(env_override).expanduser()
18
+
19
+ candidates = [
20
+ Path(".claude/project-context.json"),
21
+ Path(".claude/project-context/project-context.json"),
22
+ Path("project-context.json"),
23
+ ]
24
+
25
+ for candidate in candidates:
26
+ if candidate.is_file():
27
+ return candidate
28
+
29
+ # Fall back to the legacy expectation; load_project_context will surface error
30
+ return candidates[0]
31
+
32
+ DEFAULT_CONTEXT_PATH = get_default_context_path()
12
33
 
13
34
  # Path to provider-specific contract files
14
35
  # When running from installed project: .claude/config (symlinked)