@tgoodington/intuition 2.0.1 → 2.0.2

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.
Files changed (2) hide show
  1. package/README.md +285 -215
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -1,312 +1,378 @@
1
1
  # Intuition
2
2
 
3
- Four Claude Code skills for structured planning, execution orchestration, project initialization, and compliance enforcement.
3
+ A three-agent system for software project planning and execution. Intuition splits complex work into focused phases with clean context boundaries, dramatically improving success rates.
4
4
 
5
- Intuition provides:
6
- - **`/intuition-start`**: Load project context and enforce compliance with documented protocols
7
- - **`/intuition-plan`**: Waldo planning agent for thoughtful architecture and feature planning
8
- - **`/intuition-execute`**: Architect execution orchestrator for implementing approved plans
9
- - **`/intuition-initialize`**: Project initialization for structured knowledge system tracking decisions, bugs, and progress
5
+ **The Three Agents:**
6
+ - **Waldo** (`/intuition-discovery`) - Explores problems through GAPP discovery and Socratic dialogue
7
+ - **Magellan** (`/intuition-plan`) - Synthesizes discovery into structured, executable plans
8
+ - **Faraday** (`/intuition-execute`) - Orchestrates implementation with methodical precision
9
+
10
+ **Supporting Skills:**
11
+ - `/intuition-start` - Load project context and workflow status
12
+ - `/intuition-initialize` - Setup project memory system
10
13
 
11
14
  ## Quick Start
12
15
 
13
16
  ```bash
14
17
  # Install globally
15
- npm install -g .
18
+ npm install -g @tgoodington/intuition
16
19
 
17
20
  # In Claude Code, typical workflow:
18
- /intuition-initialize # Set up project memory (once per project)
19
- /intuition-start # Load context at start of session
20
- /intuition-plan "..." # Plan new features
21
- /intuition-execute # Execute the plan
21
+ /intuition-start # Load context and check workflow status
22
+ /intuition-discovery # Waldo guides you through GAPP discovery
23
+ /intuition-plan # Magellan creates a structured plan
24
+ /intuition-execute # Faraday orchestrates implementation
22
25
  ```
23
26
 
24
- ## Skills
27
+ ## Why Three Agents?
28
+
29
+ Each agent has a focused responsibility:
30
+
31
+ | Agent | Skill | Phase | Input | Output | Focus |
32
+ |-------|-------|-------|-------|--------|-------|
33
+ | **Waldo** | `/intuition-discovery` | Discovery | Problem description | `discovery_brief.md` | Understanding & dialogue |
34
+ | **Magellan** | `/intuition-plan` | Planning | `discovery_brief.md` | `plan.md` | Strategy & synthesis |
35
+ | **Faraday** | `/intuition-execute` | Execution | `plan.md` | Code + memory updates | Implementation & verification |
36
+
37
+ By splitting into three phases:
38
+ - **Clean context** - Each agent starts fresh, focused on one job
39
+ - **Reusable outputs** - Discovery briefs and plans are stored in project memory
40
+ - **Higher success rate** - Less context bloat = fewer dropped threads
41
+ - **Resume support** - Interrupted work picks up from last checkpoint
25
42
 
26
- ### `/intuition-start` Skill
43
+ ## The Three-Phase Workflow
27
44
 
28
- Load your project context and enforce compliance with documented protocols.
45
+ ### Phase 1: Discovery with Waldo
46
+
47
+ Explore your problem deeply using the GAPP framework:
29
48
 
30
- **Usage:**
31
49
  ```
32
- /intuition-start
50
+ /intuition-discovery
33
51
  ```
34
52
 
35
- **What happens:**
36
- 1. Reads project memory files (bugs.md, decisions.md, key_facts.md, issues.md)
37
- 2. Reads protocol documents (CLAUDE.md, AGENTS.md)
38
- 3. Summarizes project status, decisions, and constraints
39
- 4. Primes Waldo and Architect to follow project patterns
40
- 5. Offers next steps (plan, execute, or continue work)
53
+ Waldo guides you through:
54
+ - **Problem** - What's the core challenge?
55
+ - **Goals** - What does success look like?
56
+ - **UX Context** - Who will use this and how?
57
+ - **Personalization** - What drives this work for you?
58
+
59
+ **Output:** `docs/project_notes/discovery_brief.md`
41
60
 
42
- **When to use:** Start of every session to load full project context.
61
+ Uses Socratic questioning and systems thinking to surface authentic intentions before planning begins.
43
62
 
44
- ### `/intuition-plan` Skill
63
+ ### Phase 2: Planning with Magellan
45
64
 
46
- Start a planning session with Waldo, your planning thought partner.
65
+ Create a structured plan from discovery:
47
66
 
48
- **Usage:**
49
67
  ```
50
- /intuition-plan "Implement real-time notifications with WebSockets"
68
+ /intuition-plan
51
69
  ```
52
70
 
53
- **What happens:**
54
- 1. Auto-initializes project memory on first run (creates `docs/project_notes/`)
55
- 2. Waldo engages in dialogue to understand your goals
56
- 3. Waldo researches your codebase and develops a detailed plan through reflection and refinement
57
- 4. Plan is saved to `docs/project_notes/project_plan.md`
58
- 5. You can review and approve before execution
71
+ Magellan:
72
+ 1. Reads your discovery brief
73
+ 2. Researches your codebase (parallel agents)
74
+ 3. Synthesizes insights into a strategic plan
75
+ 4. Auto-detects planning depth (simple vs. complex)
76
+ 5. Presents plan for your approval
77
+
78
+ **Output:** `docs/project_notes/plan.md`
79
+
80
+ Includes tasks with acceptance criteria, dependencies, risks, and execution notes.
59
81
 
60
- ### `/intuition-execute` Skill
82
+ ### Phase 3: Execution with Faraday
61
83
 
62
- Execute an existing plan with Architect, the execution orchestrator.
84
+ Execute the approved plan:
63
85
 
64
- **Usage:**
65
86
  ```
66
87
  /intuition-execute
67
88
  ```
68
89
 
69
- **What happens:**
70
- 1. Architect reads your plan from `docs/project_notes/project_plan.md`
71
- 2. Reviews plan for feasibility and gaps
72
- 3. Creates tasks with dependencies
73
- 4. Delegates implementation to specialized sub-agents
74
- 5. Verifies outputs and reports completion
90
+ Faraday:
91
+ 1. Reads plan and discovery context
92
+ 2. Confirms approach with you
93
+ 3. Delegates to specialized sub-agents (parallel when possible)
94
+ 4. Verifies outputs against acceptance criteria
95
+ 5. Updates project memory
96
+ 6. Reports completion
75
97
 
76
- **Prerequisite:** A plan must exist from a previous `/intuition-plan` session
98
+ **Output:** Implemented code + updated project memory
77
99
 
78
- ### `/intuition-initialize` Skill
100
+ ## Project Memory System
79
101
 
80
- Initialize the project memory system.
102
+ Intuition maintains persistent project knowledge:
81
103
 
82
- **Usage:**
83
104
  ```
84
- /intuition-initialize
105
+ docs/project_notes/
106
+ ├── .project-memory-state.json (workflow status & resume data)
107
+ ├── discovery_brief.md (from Waldo)
108
+ ├── plan.md (from Magellan)
109
+ ├── bugs.md (known issues & solutions)
110
+ ├── decisions.md (architectural decisions)
111
+ ├── key_facts.md (project configuration)
112
+ └── issues.md (work log)
85
113
  ```
86
114
 
87
- **What happens:**
88
- 1. Creates `docs/project_notes/` directory structure
89
- 2. Sets up memory files:
90
- - `bugs.md` - Bug log with solutions
91
- - `decisions.md` - Architectural decisions
92
- - `key_facts.md` - Project configuration
93
- - `issues.md` - Work log
94
- 3. Configures project memory tracking
95
- 4. Sets up CLAUDE.md and AGENTS.md with memory-aware protocols
115
+ All three agents reference and update project memory for consistency.
96
116
 
97
- ## Workflow
117
+ ## Workflow Status
98
118
 
99
- A typical Intuition workflow in Claude Code:
119
+ Check where you are in the workflow:
100
120
 
101
121
  ```
102
- 1. /intuition-initialize
103
- └─ Initialize project memory system in docs/project_notes/
104
-
105
- 2. /intuition-start (at beginning of each session)
106
- └─ Load project context and enforce protocol compliance
107
- └─ Understand project decisions, constraints, and status
108
-
109
- 3. /intuition-plan "Feature description"
110
- └─ Work with Waldo to develop a comprehensive plan
111
- └─ Plan saved to docs/project_notes/project_plan.md
112
- └─ Review and approve plan
113
-
114
- 4. /intuition-execute
115
- └─ Architect reads the plan and enforces compliance
116
- └─ Delegates work to specialized sub-agents
117
- └─ Executes plan tasks with verification
118
- └─ Reports completion
119
-
120
- 5. (Iterate as needed)
121
- └─ /intuition-start (reload context for new session)
122
- └─ /intuition-plan "Next phase"
123
- └─ /intuition-execute
122
+ /intuition-start
124
123
  ```
125
124
 
126
- ## Using Intuition Skills
125
+ Returns current status:
126
+ - **Discovery:** In progress / Complete
127
+ - **Planning:** In progress / Complete / Approved
128
+ - **Execution:** In progress / Complete
127
129
 
128
- After installation via npm, the four Intuition skills are available globally in Claude Code:
130
+ Also provides helpful next-step suggestions.
129
131
 
130
- ### `/intuition-initialize` Skill
132
+ ## File Outputs
131
133
 
132
- Set up project memory and structured knowledge system.
134
+ ### discovery_brief.md (From Waldo)
133
135
 
134
- ```
135
- /intuition-initialize
136
- ```
136
+ ```markdown
137
+ # Discovery Brief
137
138
 
138
- This creates:
139
- - `docs/project_notes/bugs.md` - Bug log with solutions
140
- - `docs/project_notes/decisions.md` - Architectural decisions
141
- - `docs/project_notes/key_facts.md` - Project configuration
142
- - `docs/project_notes/issues.md` - Work log
143
- - Configures CLAUDE.md and AGENTS.md with memory protocols
139
+ ## Problem
140
+ [Core challenge and context]
144
141
 
145
- **Run this first** when starting a new project.
142
+ ## Goals
143
+ [Success criteria]
146
144
 
147
- ### `/intuition-plan` Skill
145
+ ## User Context
146
+ [Personas and workflows]
148
147
 
149
- Start a planning session with Waldo:
148
+ ## Personalization / Motivation
149
+ [What drives this work]
150
150
 
151
- ```
152
- /intuition-plan "Add user authentication to the app"
151
+ ## Scope
152
+ [In scope / out of scope]
153
+
154
+ ## Assumptions
155
+ [With confidence levels]
153
156
  ```
154
157
 
155
- Waldo will:
156
- 1. Engage in dialogue to understand your goals
157
- 2. Research your codebase for context
158
- 3. Develop a structured plan with confidence scoring
159
- 4. Reflect and refine before finalizing
160
- 5. Save the plan to `docs/project_notes/project_plan.md`
158
+ ### plan.md (From Magellan)
161
159
 
162
- ### `/intuition-execute` Skill
160
+ ```markdown
161
+ # Plan: [Title]
163
162
 
164
- Execute an existing plan with Architect:
163
+ ## Objective
164
+ [What will be accomplished]
165
165
 
166
- ```
167
- /intuition-execute
166
+ ## Discovery Summary
167
+ [Key insights from Waldo]
168
+
169
+ ## Research Context
170
+ [Codebase findings]
171
+
172
+ ## Approach
173
+ [Strategy and rationale]
174
+
175
+ ## Tasks
176
+ [With acceptance criteria, dependencies]
177
+
178
+ ## Risks & Mitigations
179
+ [Identified risks]
180
+
181
+ ## Execution Notes for Faraday
182
+ [Guidance for execution]
168
183
  ```
169
184
 
170
- Architect will:
171
- 1. Read your plan from `docs/project_notes/project_plan.md`
172
- 2. Review for feasibility and gaps
173
- 3. Create tasks with dependencies
174
- 4. Delegate to specialized sub-agents
175
- 5. Verify outputs and report completion
185
+ ## Resume Support
176
186
 
177
- **Prerequisite:** A plan must exist from a previous `/intuition-plan` session.
187
+ All three agents can resume interrupted work:
178
188
 
179
- ### Verification
189
+ - **Waldo** resumes from the last GAPP phase
190
+ - **Magellan** resumes from research or draft state
191
+ - **Faraday** resumes from the last completed task
180
192
 
181
- To verify installation in Claude Code:
193
+ Just run the skill again and it will pick up from the checkpoint.
182
194
 
183
- 1. Open any project in Claude Code
184
- 2. Type `/` to see available skills
185
- 3. You should see `/intuition-start`, `/intuition-plan`, `/intuition-execute`, and `/intuition-initialize`
195
+ ## Discovery Revision
186
196
 
187
- If skills are not available:
188
- - Ensure installation completed (check `~/.claude/skills/` contains the skill directories)
189
- - Restart Claude Code if just installed
190
- - See [INSTALLATION.md](./INSTALLATION.md) for troubleshooting
197
+ If you need to revise your discovery:
191
198
 
192
- ## Project Structure
199
+ 1. Run `/intuition-discovery` again
200
+ 2. Waldo updates `discovery_brief.md`
201
+ 3. Magellan detects the change and offers to re-plan
202
+ 4. New plan created with updated context
203
+
204
+ ## Installation
193
205
 
206
+ ### Install Globally (Recommended)
207
+
208
+ ```bash
209
+ npm install -g @tgoodington/intuition
194
210
  ```
195
- intuition/
196
- ├── skills/
197
- │ ├── intuition-start/ # Context loading and compliance enforcement
198
- │ ├── intuition-plan/ # Planning with Waldo
199
- │ ├── intuition-execute/ # Execution with Architect
200
- │ └── intuition-initialize/ # Project memory setup
201
- ├── agents/
202
- │ ├── waldo.md # Planning agent definition
203
- │ ├── architect.md # Execution orchestrator
204
- │ └── [other sub-agents]
205
- ├── scripts/
206
- │ └── install-skills.js # Installation script
207
- ├── package.json # npm package config
208
- └── README.md # This file
211
+
212
+ This installs five skills globally to `~/.claude/skills/`:
213
+ - `/intuition-start` - Load project context
214
+ - `/intuition-initialize` - Setup project memory
215
+ - `/intuition-discovery` - Waldo's discovery
216
+ - `/intuition-plan` - Magellan's planning
217
+ - `/intuition-execute` - Faraday's execution
218
+
219
+ ### Install from Source (Development)
220
+
221
+ ```bash
222
+ cd intuition
223
+ npm install -g .
209
224
  ```
210
225
 
211
- ## Agents
226
+ ### Verify Installation
212
227
 
213
- ### Waldo - Planning Thought Partner
228
+ In Claude Code, type `/` to see available skills. You should see all five:
229
+ - `/intuition-start`
230
+ - `/intuition-initialize`
231
+ - `/intuition-discovery`
232
+ - `/intuition-plan`
233
+ - `/intuition-execute`
214
234
 
215
- Waldo specializes in collaborative planning. When you run `intuition plan`, Waldo:
216
- - Engages in dialogue to understand your goals
217
- - Researches your codebase for context
218
- - Develops structured plans with confidence scoring
219
- - Reflects and refines before finalizing
220
- - Outputs markdown plans for review
235
+ ## Skills Reference
221
236
 
222
- Read more in `agents/waldo.md`
237
+ ### `/intuition-start`
223
238
 
224
- ### Architect - Execution Orchestrator
239
+ Load project context and check workflow status.
225
240
 
226
- Architect specializes in execution coordination. When you run `intuition execute`, Architect:
227
- - Reviews the plan for completeness
228
- - Breaks plans into discrete tasks
229
- - Delegates to specialized sub-agents (Code Writer, Test Runner, etc.)
230
- - Verifies outputs against acceptance criteria
231
- - Handles failures with retry/fallback strategies
241
+ ```
242
+ /intuition-start
243
+ ```
232
244
 
233
- Read more in `agents/architect.md`
245
+ **Does:**
246
+ - Loads project memory files
247
+ - Checks workflow status
248
+ - Suggests next steps based on status
249
+ - Enforces project protocols
234
250
 
235
- ## Project Memory System
251
+ **When to use:** Start of every session
236
252
 
237
- Intuition includes a project memory system that maintains institutional knowledge across sessions.
253
+ ### `/intuition-initialize`
238
254
 
239
- **Memory files:**
240
- - `docs/project_notes/bugs.md` - Known bugs and solutions
241
- - `docs/project_notes/decisions.md` - Architectural decisions
242
- - `docs/project_notes/key_facts.md` - Project configuration
243
- - `docs/project_notes/issues.md` - Work log and tickets
255
+ Setup project memory system.
244
256
 
245
- Memory is automatically referenced by Waldo and Architect to maintain consistency and avoid repeating past mistakes.
257
+ ```
258
+ /intuition-initialize
259
+ ```
246
260
 
247
- See `skills/intuition-initialize/SKILL.md` for full details.
261
+ **Creates:**
262
+ - `docs/project_notes/` directory
263
+ - `bugs.md` - Bug tracking
264
+ - `decisions.md` - Architecture decisions
265
+ - `key_facts.md` - Project configuration
266
+ - `issues.md` - Work log
267
+ - `.project-memory-state.json` - Workflow tracking
248
268
 
249
- ## Installation
269
+ **When to use:** Once per project, at the start
250
270
 
251
- ### Install Globally via npm (Recommended)
271
+ ### `/intuition-discovery`
252
272
 
253
- ```bash
254
- npm install -g intuition
273
+ Explore your problem with Waldo.
274
+
275
+ ```
276
+ /intuition-discovery
255
277
  ```
256
278
 
257
- This automatically installs the four Intuition skills to `~/.claude/skills/` for use across all projects with Claude Code:
258
- - `/intuition-start` - Load project context
259
- - `/intuition-plan` - Planning with Waldo
260
- - `/intuition-execute` - Execution with Architect
261
- - `/intuition-initialize` - Project initialization
279
+ **Waldo does:**
280
+ - GAPP discovery (Problem Goals → UX Context → Personalization)
281
+ - Socratic questioning
282
+ - Systems thinking perspective
283
+ - Clarifying questions to validate understanding
284
+ - Saves `discovery_brief.md`
285
+
286
+ **When to use:** Start of new work
287
+
288
+ ### `/intuition-plan`
289
+
290
+ Create a plan with Magellan.
262
291
 
263
- **Verification:**
264
- ```bash
265
- ls ~/.claude/skills/intuition-*
292
+ ```
293
+ /intuition-plan
266
294
  ```
267
295
 
268
- ### Install from Source (Development)
296
+ **Magellan does:**
297
+ - Reads your discovery brief
298
+ - Researches codebase (parallel agents)
299
+ - Synthesizes into strategic plan
300
+ - Auto-detects planning depth
301
+ - Saves `plan.md`
269
302
 
270
- ```bash
271
- cd intuition
272
- npm install -g .
303
+ **When to use:** After discovery is complete
304
+
305
+ ### `/intuition-execute`
306
+
307
+ Execute the plan with Faraday.
308
+
309
+ ```
310
+ /intuition-execute
273
311
  ```
274
312
 
275
- After installation, restart Claude Code and the skills will be available globally.
313
+ **Faraday does:**
314
+ - Reads plan and discovery context
315
+ - Confirms approach with you
316
+ - Delegates to sub-agents (parallel when possible)
317
+ - Verifies outputs
318
+ - Updates project memory
276
319
 
277
- ## Requirements
320
+ **When to use:** After plan is approved
278
321
 
279
- - Node.js 14.0.0 or higher
280
- - Access to Claude Code or similar agent system
322
+ ## Sub-Agents
281
323
 
282
- ## Architecture
324
+ Faraday coordinates these specialized agents:
283
325
 
284
- Intuition provides four integrated skills for Claude Code:
326
+ | Agent | Purpose |
327
+ |-------|---------|
328
+ | Code Writer | Implementation |
329
+ | Test Runner | Testing & verification |
330
+ | Code Reviewer | Quality review |
331
+ | Documentation | Updates docs |
332
+ | Research | Codebase exploration |
333
+ | Security Expert | Vulnerability scanning |
334
+ | Technical Spec Writer | Specification creation |
335
+ | Communications Specialist | User-facing documentation |
285
336
 
286
- 1. **Context Loading** (`/intuition-start`) - Loads project memory and enforces protocol compliance
287
- 2. **Initialization** (`/intuition-initialize`) - Sets up project memory and knowledge system
288
- 3. **Planning** (`/intuition-plan` with Waldo agent) - Structured planning with reflection and dialogue
289
- 4. **Execution** (`/intuition-execute` with Architect agent) - Orchestration and verification through sub-agents
337
+ ## Documentation
290
338
 
291
- All skills coordinate through the project memory system (`docs/project_notes/`) for consistency and context across sessions. `/intuition-start` ensures every session begins with full project context and compliance enforcement.
339
+ - **Workflow Guide:** `docs/intuition-workflow.md`
340
+ - **Architecture:** `docs/intuition-architecture.md`
341
+ - **Waldo Reference:** `skills/intuition-discovery/references/waldo_core.md`
342
+ - **Magellan Reference:** `skills/intuition-plan/references/magellan_core.md`
343
+ - **Faraday Reference:** `skills/intuition-execute/references/faraday_core.md`
292
344
 
293
- ### Sub-Agent System
345
+ ## Requirements
346
+
347
+ - Node.js 14.0.0 or higher
348
+ - Claude Code or similar agent system
349
+
350
+ ## Project Structure
294
351
 
295
- When `/execute` runs, Architect delegates to specialized sub-agents:
296
- - Code Writer - Implementation
297
- - Test Runner - Verification
298
- - Code Reviewer - Quality assurance
299
- - Documentation - Knowledge updates
300
- - Research - Investigation and exploration
301
- - Security Expert - Vulnerability detection
352
+ ```
353
+ intuition/
354
+ ├── skills/
355
+ │ ├── intuition-start/ # Context loading & workflow status
356
+ │ ├── intuition-initialize/ # Project memory setup
357
+ │ ├── intuition-discovery/ # Waldo - discovery & GAPP
358
+ │ ├── intuition-plan/ # Magellan - strategic planning
359
+ │ └── intuition-execute/ # Faraday - orchestrated execution
360
+ ├── agents/ # Agent definitions
361
+ ├── docs/ # Documentation
362
+ │ ├── intuition-workflow.md # User guide
363
+ │ └── intuition-architecture.md # Technical details
364
+ ├── scripts/ # Installation scripts
365
+ ├── package.json # npm package config
366
+ └── README.md # This file
367
+ ```
302
368
 
303
369
  ## Contributing
304
370
 
305
371
  Contributions welcome! Areas for enhancement:
306
372
  - Additional specialized sub-agents
307
- - Plan format improvements
308
- - Memory system enhancements
309
- - CLI features
373
+ - Workflow improvements
374
+ - Documentation enhancements
375
+ - Framework extensions
310
376
 
311
377
  ## License
312
378
 
@@ -315,15 +381,19 @@ MIT
315
381
  ## Support
316
382
 
317
383
  For issues or questions:
318
- - Check `docs/project_notes/` for known issues and decisions
319
- - Review agent definitions in `agents/`
320
- - Check the skill documentation in `skills/`
321
-
322
- ## See Also
323
-
324
- - Intuition Start Skill: `skills/intuition-start/SKILL.md`
325
- - Intuition Plan Skill: `skills/intuition-plan/SKILL.md`
326
- - Intuition Execute Skill: `skills/intuition-execute/SKILL.md`
327
- - Intuition Initialize Skill: `skills/intuition-initialize/SKILL.md`
328
- - Waldo Agent: `agents/waldo.md`
329
- - Architect Agent: `agents/architect.md`
384
+ - Check `docs/intuition-workflow.md` for user guide
385
+ - Check `docs/intuition-architecture.md` for technical details
386
+ - Review skill documentation in `skills/*/SKILL.md`
387
+ - Check project memory in `docs/project_notes/` for decisions and issues
388
+
389
+ ## Version History
390
+
391
+ ### 2.0.0+
392
+ - Complete refactor into three-agent system (Waldo → Magellan → Faraday)
393
+ - GAPP framework for discovery
394
+ - File-based handoffs through project memory
395
+ - Enhanced state management
396
+ - Resume support for all phases
397
+
398
+ ### 1.x
399
+ - Original monolithic planning system
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@tgoodington/intuition",
3
- "version": "2.0.1",
3
+ "version": "2.0.2",
4
4
  "description": "Three-agent system for software project planning and execution. Waldo (discovery), Magellan (planning), Faraday (execution) with file-based handoffs through project memory.",
5
5
  "keywords": [
6
6
  "claude-code",