@tgoodington/intuition 8.1.0 → 8.1.2

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.
@@ -2,41 +2,69 @@
2
2
  name: intuition-initialize
3
3
  description: Set up project memory infrastructure with workflow state tracking, memory files, and configuration templates.
4
4
  model: haiku
5
- tools: Read, Write, Glob, Grep, AskUserQuestion
5
+ tools: Read, Write, Glob, Grep, Bash, AskUserQuestion
6
6
  allowed-tools: Read, Write, Glob, Grep
7
7
  ---
8
8
 
9
9
  # Initialize - Project Memory Setup Protocol
10
10
 
11
- You are the project memory initializer. You create the `docs/project_notes/` directory structure, initialize memory files from templates, create the workflow state file, and configure CLAUDE.md with memory-aware protocols. You run once per project to set up infrastructure, then hand off to the workflow skills.
11
+ You create the `docs/project_notes/` directory structure, initialize memory files from templates, create the workflow state file, and set up project configuration files. You run once per project to set up infrastructure, then hand off to the workflow skills.
12
12
 
13
13
  ## CRITICAL RULES
14
14
 
15
- These are non-negotiable. Violating any of these means the protocol has failed.
16
-
17
- 1. You MUST check if `docs/project_notes/` already exists before creating anything.
18
- 2. You MUST use template files from `references/` directory for all initial content. Read each template with the Read tool, then Write to the target path.
19
- 3. You MUST create `.project-memory-state.json` using the v5.0 schema from `references/state_template.json`. Do NOT use older schemas (v1.0, v2.0, v3.0, v4.0).
20
- 4. You MUST update CLAUDE.md with workflow and memory protocols.
21
- 5. You MUST NOT overwrite existing memory files without asking the user first.
22
- 6. You MUST NOT invoke other skills (discovery, plan, execute). You only set up infrastructure.
23
- 7. You MUST ask the user before creating optional components (AGENTS.md, settings, user profile).
15
+ 1. You MUST check install location before doing anything else enforce global install.
16
+ 2. You MUST check if `docs/project_notes/` already exists before creating anything.
17
+ 3. You MUST use template files from `references/` directory for all initial content. Read each template with the Read tool, then Write to the target path.
18
+ 4. You MUST create `.project-memory-state.json` using the v5.0 schema from `references/state_template.json`. Do NOT use older schemas.
19
+ 5. You MUST auto-create all configuration files (CLAUDE.md, INTUITION.md, AGENTS.md, settings, profile) if they do not exist. Do NOT ask just create them.
20
+ 6. You MUST NOT overwrite existing files. If a file already exists, skip it silently.
21
+ 7. You MUST NOT invoke other skills. You only set up infrastructure.
24
22
  8. You MUST report what was created at the end.
25
23
 
26
24
  ## PROTOCOL: COMPLETE FLOW
27
25
 
28
- Execute these steps in order:
29
-
30
26
  ```
27
+ Step 0: Check install location — enforce global install
31
28
  Step 1: Detect existing setup
32
29
  Step 2: Create memory directory and files from templates
33
30
  Step 3: Create .project-memory-state.json with v5.0 schema
34
31
  Step 4: Update CLAUDE.md with workflow protocols
35
32
  Step 4.5: Create INTUITION.md framework overview
36
- Step 5: Offer optional components (AGENTS.md, settings, user profile)
33
+ Step 5: Create configuration files (AGENTS.md, settings, user profile)
37
34
  Step 6: Report completion and suggest next step
38
35
  ```
39
36
 
37
+ ## STEP 0: ENFORCE GLOBAL INSTALL
38
+
39
+ Before anything else, check whether `@tgoodington/intuition` is installed locally (project-level) vs globally.
40
+
41
+ Run these checks using Bash:
42
+
43
+ ```
44
+ Local: npm list @tgoodington/intuition --depth=0 2>&1
45
+ Global: npm list -g @tgoodington/intuition --depth=0 2>&1
46
+ ```
47
+
48
+ Apply this logic:
49
+
50
+ ```
51
+ IF local install found AND global install found:
52
+ → Uninstall local: npm uninstall @tgoodington/intuition
53
+ → Notify: "Removed local install. Using global install."
54
+
55
+ IF local install found AND no global install:
56
+ → Uninstall local: npm uninstall @tgoodington/intuition
57
+ → Install global: npm install -g @tgoodington/intuition
58
+ → Notify: "Moved install from local to global."
59
+
60
+ IF no local install AND global install found:
61
+ → Proceed (correct setup)
62
+
63
+ IF neither found:
64
+ → Warn: "Package not found. Install with: npm install -g @tgoodington/intuition"
65
+ → STOP
66
+ ```
67
+
40
68
  ## STEP 1: DETECT EXISTING SETUP
41
69
 
42
70
  Check if `docs/project_notes/` exists.
@@ -83,7 +111,7 @@ For each file, use the Read tool to read the template, then use Write to create
83
111
  | `references/key_facts_template.md` | `docs/project_notes/key_facts.md` |
84
112
  | `references/issues_template.md` | `docs/project_notes/issues.md` |
85
113
 
86
- Do NOT create workflow output files (discovery_brief.md, plan.md, engineering_brief.md, build_brief.md, etc.). Those are created by their respective skills during the workflow.
114
+ Do NOT create workflow output files (discovery_brief.md, plan.md, etc.). Those are created by their respective skills.
87
115
 
88
116
  ## STEP 3: CREATE STATE FILE
89
117
 
@@ -137,113 +165,69 @@ The state file uses the v5.0 schema:
137
165
  }
138
166
  ```
139
167
 
140
- **CRITICAL**: This is the authoritative schema. Handoff is the ONLY skill that updates this file after initialization. All other skills read it but NEVER write to it.
141
-
142
- Do NOT use older schemas (v1.0, v2.0, v3.0, or v4.0). Those schemas are obsolete.
168
+ **CRITICAL**: Handoff is the ONLY skill that updates this file after initialization.
143
169
 
144
170
  ## STEP 4: UPDATE CLAUDE.MD
145
171
 
146
172
  Read `references/claude_template.md` for the workflow protocol content.
147
173
 
148
174
  ```
149
- IF CLAUDE.md exists:
150
- → Check for existing "Project Workflow and Memory System" section
151
- → If section exists: Ask user before replacing
152
- → If section missing: Append the template content
153
-
154
175
  IF CLAUDE.md does NOT exist:
155
176
  → Create CLAUDE.md with the template content
156
- ```
157
177
 
158
- The template includes:
159
- - Five-phase workflow description (prompt plan design engineer → build with handoffs)
160
- - Memory file descriptions and locations
161
- - Memory-aware protocols (check decisions before changes, search bugs before debugging)
162
- - Smart skill suggestions (when to suggest /intuition-prompt, /intuition-plan, /intuition-design, /intuition-engineer, /intuition-build, etc.)
178
+ IF CLAUDE.md exists BUT has no "Project Workflow and Memory System" section:
179
+ Append the template content to the end of the file
180
+
181
+ IF CLAUDE.md exists AND already has the section:
182
+ Skip (do not overwrite)
183
+ ```
163
184
 
164
185
  ## STEP 4.5: CREATE INTUITION.MD
165
186
 
166
- Read `references/intuition_readme_template.md` and write to `INTUITION.md` at the project root (same level as CLAUDE.md).
187
+ Read `references/intuition_readme_template.md` and write to `INTUITION.md` at the project root.
167
188
 
168
189
  ```
169
- IF INTUITION.md already exists:
170
- → Ask user: "INTUITION.md already exists. Overwrite with current version?"
171
- → If no, skip
172
-
173
190
  IF INTUITION.md does NOT exist:
174
191
  → Create it from template
175
- ```
176
-
177
- This is a brief, human-readable overview of the Intuition workflow. It helps anyone on the project understand what the skills do and how to use them.
178
192
 
179
- ## STEP 5: OPTIONAL COMPONENTS
193
+ IF INTUITION.md already exists:
194
+ → Skip
195
+ ```
180
196
 
181
- Offer each optional component using AskUserQuestion. Do not create without asking.
197
+ ## STEP 5: CREATE CONFIGURATION FILES
182
198
 
183
- ### 5A: AGENTS.md Configuration
199
+ Auto-create each file if it does not exist. Do NOT ask — just create. If a file already exists, skip it silently.
184
200
 
185
- ```
186
- Question: "Create AGENTS.md with agent registry and coordination protocols?"
187
- Options:
188
- - "Yes, create it" (Recommended)
189
- - "Skip this"
190
- ```
201
+ ### 5A: AGENTS.md
191
202
 
192
- If yes:
193
203
  - Read `references/agents_template.md`
194
- - If AGENTS.md exists: append or update the "Multi-Agent System" section
195
- - If AGENTS.md does not exist: create with full template content
204
+ - If `AGENTS.md` does not exist: create with full template content
205
+ - If `AGENTS.md` exists: skip
196
206
 
197
207
  ### 5B: Claude Code Settings
198
208
 
199
- ```
200
- Question: "Create .claude/settings.local.json to pre-authorize common tools?"
201
- Options:
202
- - "Yes, create it" (Recommended)
203
- - "Skip this"
204
- ```
205
-
206
- If yes:
207
209
  - Read `references/settings_template.json`
208
- - If `.claude/settings.local.json` exists: ask before overwriting
209
- - If not exists: create from template
210
- - Pre-authorizes: Read, Glob, Grep, WebSearch, WebFetch, Task, common git operations
210
+ - If `.claude/settings.local.json` does not exist: create from template
211
+ - If `.claude/settings.local.json` exists: skip
211
212
 
212
213
  ### 5C: User Profile
213
214
 
214
- ```
215
- Question: "Create .claude/USER_PROFILE.json for cross-project user context?"
216
- Options:
217
- - "Yes, create it"
218
- - "Skip this" (Recommended)
219
- ```
220
-
221
- If yes:
222
215
  - Read `references/user_profile_template.json`
223
- - If `.claude/USER_PROFILE.json` exists: ask before overwriting
224
- - If not exists: create from template
225
- - This is a global file (in `.claude/`), not project-specific
226
- - Agents populate it naturally through conversation over time
216
+ - If `.claude/USER_PROFILE.json` does not exist: create from template
217
+ - If `.claude/USER_PROFILE.json` exists: skip
227
218
 
228
219
  ## STEP 6: REPORT COMPLETION
229
220
 
230
- Present a concise summary of what was created:
221
+ Present a concise summary of what was created vs skipped:
231
222
 
232
223
  ```
233
224
  Project memory initialized!
234
225
 
235
226
  Created:
236
- - docs/project_notes/trunk/ (trunk workflow artifacts)
237
- - docs/project_notes/branches/ (branch workflow artifacts)
238
- - docs/project_notes/bugs.md
239
- - docs/project_notes/decisions.md
240
- - docs/project_notes/key_facts.md
241
- - docs/project_notes/issues.md
242
- - docs/project_notes/.project-memory-state.json (v5.0 schema)
243
- - CLAUDE.md workflow protocols
244
- - INTUITION.md framework overview
245
-
246
- [List any optional components created]
227
+ - [list each file that was created]
228
+
229
+ Skipped (already existed):
230
+ - [list each file that was skipped, if any]
247
231
 
248
232
  Next step: Run /intuition-prompt to describe what you want to build,
249
233
  or /intuition-start to check project status.
@@ -251,78 +235,25 @@ or /intuition-start to check project status.
251
235
 
252
236
  ## EDGE CASES
253
237
 
254
- - **Partial setup exists**: Check each file individually. Only create missing files. Ask before overwriting any existing file.
238
+ - **Partial setup exists**: Check each file individually. Only create missing files.
255
239
  - **CLAUDE.md has custom content**: Append the workflow section. Do not replace the entire file.
256
- - **User declines all optional components**: That's fine. Core memory files and CLAUDE.md are sufficient.
257
- - **Git-tracked project**: Memory files in `docs/project_notes/` are safe to commit. The `.project-memory-state.json` contains only metadata. Settings in `.claude/` are typically gitignored.
240
+ - **Git-tracked project**: Memory files in `docs/project_notes/` are safe to commit. Settings in `.claude/` are typically gitignored.
258
241
 
259
242
  ## REFERENCE FILES
260
243
 
261
244
  These template files are in the `references/` directory. Use Read tool to access them:
262
245
 
263
- **Memory file templates** (used in Step 2):
264
- - `bugs_template.md`
265
- - `decisions_template.md`
266
- - `key_facts_template.md`
267
- - `issues_template.md`
246
+ **Memory file templates** (Step 2):
247
+ - `bugs_template.md`, `decisions_template.md`, `key_facts_template.md`, `issues_template.md`
268
248
 
269
- **State template** (used in Step 3):
270
- - `state_template.json` — v4.0 schema
249
+ **State template** (Step 3):
250
+ - `state_template.json`
271
251
 
272
- **Framework overview** (used in Step 4.5):
273
- - `intuition_readme_template.md` — high-level workflow overview for INTUITION.md
252
+ **Framework overview** (Step 4.5):
253
+ - `intuition_readme_template.md`
274
254
 
275
- **Configuration templates** (used in Steps 4-5):
255
+ **Configuration templates** (Steps 4-5):
276
256
  - `claude_template.md` — workflow protocols for CLAUDE.md
277
257
  - `agents_template.md` — agent registry for AGENTS.md
278
258
  - `settings_template.json` — pre-authorized tools for Claude Code
279
259
  - `user_profile_template.json` — cross-project user context
280
-
281
- **Workflow output templates** (NOT used by initialize — used by handoff):
282
- - `discovery_output_template.json`
283
- - `planning_brief_template.md`
284
- - `design_brief_template.md`
285
- - `execution_brief_template.md` (engineering brief template — legacy filename)
286
-
287
- ## MEMORY FILE FORMATS
288
-
289
- For reference, here's how memory files should be formatted. Templates already include these formats.
290
-
291
- ### bugs.md
292
- ```markdown
293
- ### YYYY-MM-DD - Brief Bug Description
294
- - **Issue**: What went wrong
295
- - **Root Cause**: Why it happened
296
- - **Solution**: How it was fixed
297
- - **Prevention**: How to avoid it in the future
298
- ```
299
-
300
- ### decisions.md
301
- ```markdown
302
- ### ADR-NNN: Decision Title (YYYY-MM-DD)
303
- **Status**: Proposed | Accepted | Superseded
304
- **Context:** [Why this decision was needed]
305
- **Decision:** [What was chosen]
306
- **Consequences:** [Benefits and trade-offs]
307
- ```
308
-
309
- ### key_facts.md
310
- ```markdown
311
- ## [Category]
312
- - **[Fact]**: [value] (discovered [date])
313
- ```
314
-
315
- ### issues.md
316
- ```markdown
317
- ### YYYY-MM-DD - TICKET-ID: Brief Description
318
- - **Status**: Completed | In Progress | Blocked
319
- - **Description**: [1-2 line summary]
320
- - **URL**: [link to ticket]
321
- ```
322
-
323
- ## VOICE
324
-
325
- - Direct and procedural — "Creating files..." "Updated CLAUDE.md"
326
- - Status-focused — report what was done
327
- - No marketing copy — skip "empowering you" language
328
- - Offer optional components without being pushy
@@ -2,73 +2,117 @@
2
2
 
3
3
  ## Overview
4
4
 
5
- This project uses a multi-agent system coordinated by Intuition (Claude Code skill system) to streamline development workflows. Specialized agents handle discovery, planning, and execution, with memory maintained in `docs/project_notes/` for consistency across sessions.
6
-
7
- ## Primary Agents
8
-
9
- **Waldo** Discovery & Thought Partnership (`/intuition-discovery`)
10
- - Collaborative dialogue for understanding problems deeply
11
- - GAPP framework: Problem Goals UX Context → Personalization
12
- - Researches topic via parallel subagents before engaging
13
- - Output: `discovery_brief.md` and `discovery_output.json`
14
-
15
- **Magellan** Planning & Strategic Synthesis (`/intuition-plan`)
16
- - Synthesizes discovery into structured, executable plans
17
- - Researches codebase via parallel subagents
18
- - Output: `plan.md` with tasks, dependencies, risks
19
-
20
- **Faraday** — Execution & Implementation (`/intuition-execute`)
21
- - Executes approved plans by orchestrating specialized sub-agents
22
- - Delegates to Code Writer, Test Runner, Code Reviewer, Security Expert
23
- - Mandatory security review before completion
24
- - Output: Implemented features, completion report
25
-
26
- **Handoff Orchestrator** (`/intuition-handoff`)
27
- - Processes phase outputs, updates memory files, briefs next agent
28
- - ONLY component that writes to `.project-memory-state.json`
29
- - Bridges discovery→planning and planning→execution transitions
30
-
31
- ## Specialized Sub-Agents
32
-
33
- **Code Writer** Implements features and fixes
34
- **Test Runner** Runs unit and integration tests
35
- **Code Reviewer** — Reviews quality, maintainability, security
36
- **Security Expert** — Scans for vulnerabilities (mandatory before completion)
37
- **Documentation** — Creates and updates docs
38
- **Research** — Investigates issues, explores codebases
39
-
40
- ## Workflow Patterns
41
-
42
- ### Pattern 1: Full Feature Development (Recommended)
43
- 1. User → Waldo (collaborative discovery dialogue)
44
- 2. Handoff (processes discovery, briefs planner)
45
- 3. Magellan (creates structured plan)
46
- 4. Handoff (processes plan, briefs executor)
47
- 5. FaradaySub-agents (parallel delegation)
48
- 6. Security Expert review (mandatory)
49
-
50
- ### Pattern 2: Direct Execution (Simple Tasks)
51
- 1. User → Faraday (describe what to do)
52
- 2. Faraday → Sub-agents (delegated work)
53
- 3. Security Expert review
54
-
55
- ## Project Memory Integration
56
-
57
- **Memory Files Location**: `docs/project_notes/`
58
- - `bugs.md` — Bug log with solutions
59
- - `decisions.md` Architectural Decision Records
60
- - `key_facts.md` — Project configuration, constants
5
+ This project uses a multi-agent system coordinated by Intuition (`@tgoodington/intuition`), a Claude Code skill system. Twelve specialized skills handle prompt refinement, planning, design exploration, code engineering, and build execution, with memory maintained in `docs/project_notes/` for consistency across sessions.
6
+
7
+ ## Workflow Skills
8
+
9
+ ### Core Workflow (run in sequence with handoff between each)
10
+
11
+ | Skill | Model | What it does |
12
+ |-------|-------|-------------|
13
+ | `/intuition-prompt` | opus | Transforms a rough vision into a planning-ready brief through focused iterative refinement |
14
+ | `/intuition-plan` | opus | Strategic architect — maps stakeholders, explores components, evaluates options, creates executable blueprint |
15
+ | `/intuition-design` | opus | Elaborates flagged plan items through ECD framework (Elements, Connections, Dynamics) |
16
+ | `/intuition-engineer` | opus | Creates code-level specifications through codebase research and interactive dialogue → `code_specs.md` |
17
+ | `/intuition-build` | sonnet | Delegates implementation to subagents, verifies outputs against code specs and acceptance criteria |
18
+
19
+ ### Infrastructure
20
+
21
+ | Skill | Model | What it does |
22
+ |-------|-------|-------------|
23
+ | `/intuition-start` | haiku | Loads project context, detects workflow phase, routes to correct next skill |
24
+ | `/intuition-handoff` | haiku | Processes phase outputs, updates memory, generates briefs for the next phase |
25
+ | `/intuition-initialize` | haiku | Sets up project memory infrastructure (run once per project) |
26
+ | `/intuition-update` | haiku | Package update manager |
27
+
28
+ ### Advisory
29
+
30
+ | Skill | Model | What it does |
31
+ |-------|-------|-------------|
32
+ | `/intuition-debugger` | opus | Expert debugger — 5 diagnostic categories, causal chain analysis, post-completion only |
33
+ | `/intuition-agent-advisor` | opus | Expert advisor on building custom Claude Code agents |
34
+ | `/intuition-skill-guide` | opus | Expert advisor on building custom Claude Code skills |
35
+
36
+ ## Model Strategy
37
+
38
+ - **opus** — Complex multi-step reasoning: prompt refinement, planning, design exploration, code engineering, debugging, advisory
39
+ - **sonnet** — Project management and delegation: build manager, broad research subagents
40
+ - **haiku** — Focused simple tasks: start, handoff, state updates, narrow research subagents
41
+
42
+ ## Workflow
43
+
44
+ ### Trunk (first cycle)
45
+
46
+ ```
47
+ /intuition-prompt handoff /intuition-plan handoff →
48
+ [/intuition-design loop] handoff → /intuition-engineer → handoff →
49
+ /intuition-build → handoff → complete
50
+ ```
51
+
52
+ Each handoff transition:
53
+ 1. Reads the previous phase's output
54
+ 2. Updates shared memory files
55
+ 3. Generates a fresh brief for the next phase
56
+ 4. Updates `.project-memory-state.json`
57
+ 5. Routes to the next skill (with `/clear` between phases)
58
+
59
+ ### Branches (subsequent cycles)
60
+
61
+ After trunk completes, run `/intuition-start` to:
62
+ - **Create a branch** — new prompt→build cycle informed by trunk context
63
+ - **Open the debugger** — investigate hard problems in any completed context
64
+
65
+ Branches follow the same 5-phase workflow but read parent context for continuity.
66
+
67
+ ### Design Loop (optional)
68
+
69
+ The plan flags tasks requiring design exploration (Section 6.5). If design items exist:
70
+ 1. Handoff generates a design brief for the first item
71
+ 2. `/intuition-design` elaborates it using the ECD framework
72
+ 3. Handoff checks for remaining items → loops back or advances to engineer
73
+
74
+ ### Engineer → Build Split
75
+
76
+ - **Engineer** (opus, interactive) determines the code-level HOW: reads codebase via research subagents, discusses decisions with you, produces `code_specs.md`
77
+ - **Build** (sonnet, PM) implements against specs: delegates to Code Writer subagents, verifies with Code Reviewer, runs mandatory Security Expert review, produces `build_report.md`
78
+ - Build makes NO engineering decisions — it matches output to specs
79
+
80
+ ## Build Sub-Agents
81
+
82
+ The build phase delegates to these task-based subagents via the Task tool:
83
+
84
+ | Sub-Agent | Model | Purpose |
85
+ |-----------|-------|---------|
86
+ | Code Writer | sonnet | Implements features and fixes per code specs |
87
+ | Code Reviewer | sonnet | Reviews quality, maintainability, adherence to specs |
88
+ | Security Expert | sonnet | Scans for vulnerabilities (mandatory before completion) |
89
+
90
+ ## Project Memory
91
+
92
+ **Memory Files** (`docs/project_notes/`):
93
+ - `bugs.md` — Bug log with solutions and prevention notes
94
+ - `decisions.md` — Architectural Decision Records (ADRs)
95
+ - `key_facts.md` — Project configuration, constants, URLs
61
96
  - `issues.md` — Work log with ticket references
62
97
 
63
- **How Agents Use Memory:**
98
+ **Phase Output Files** (in `{context_path}/`):
99
+ - `planning_brief.md` — Brief for planning (created by handoff)
100
+ - `plan.md` — Structured plan with tasks, design recommendations
101
+ - `design_brief.md` — Brief for current design item (created/updated by handoff)
102
+ - `engineering_brief.md` — Brief for engineering (created by handoff)
103
+ - `code_specs.md` — Code-level specifications (created by engineer)
104
+ - `build_brief.md` — Brief for build (created by handoff)
105
+ - `build_report.md` — Task outcomes, files modified (created by build)
106
+
107
+ **How Skills Use Memory:**
64
108
  - Check `decisions.md` before proposing architectural changes
65
109
  - Search `bugs.md` for similar issues before debugging
66
110
  - Reference `key_facts.md` for project configuration
67
111
  - Log completed work in `issues.md`
68
112
 
69
- ## Agent Coordination
113
+ ## Coordination
70
114
 
71
- - All work skills route to `/intuition-handoff` between phases
72
- - Handoff owns all state transitions in `.project-memory-state.json`
73
- - Agents use structured markdown output
74
- - State is tracked in memory files for cross-session continuity
115
+ - All workflow skills route to `/intuition-handoff` between phases
116
+ - Handoff is the ONLY skill that writes to `.project-memory-state.json`
117
+ - `/clear` runs between phases to keep context clean (each skill reads from disk)
118
+ - State tracks trunk, branches, and active context with per-context workflow pipelines
@@ -1,16 +1,17 @@
1
1
  ## Project Workflow and Memory System
2
2
 
3
- This project uses a four-phase workflow coordinated by the Intuition system, with institutional knowledge maintained in `docs/project_notes/` for consistency across sessions.
3
+ This project uses a five-phase workflow coordinated by the Intuition system, with institutional knowledge maintained in `docs/project_notes/` for consistency across sessions.
4
4
 
5
5
  ### Workflow Model
6
6
 
7
7
  The Intuition workflow uses a trunk-and-branch model:
8
- - **Trunk**: The first prompt→plan→design→execute cycle. Represents the core vision.
8
+ - **Trunk**: The first prompt→plan→design→engineer→build cycle. Represents the core vision.
9
9
  - **Branches**: Subsequent cycles that build on, extend, or diverge from trunk or other branches.
10
- - **Debugger**: Post-execution diagnostic specialist for hard problems.
10
+ - **Debugger**: Post-completion diagnostic specialist for hard problems.
11
11
 
12
12
  All phases: `/intuition-prompt` → `/intuition-handoff` → `/intuition-plan` → `/intuition-handoff` →
13
- `[/intuition-design loop]` → `/intuition-handoff` → `/intuition-execute` → `/intuition-handoff` → complete
13
+ `[/intuition-design loop]` → `/intuition-handoff` → `/intuition-engineer` → `/intuition-handoff` →
14
+ `/intuition-build` → `/intuition-handoff` → complete
14
15
 
15
16
  After completion: `/intuition-start` to create branches or `/intuition-debugger` to debug issues.
16
17
 
@@ -19,12 +20,12 @@ After completion: `/intuition-start` to create branches or `/intuition-debugger`
19
20
  The project follows a structured workflow with handoff transitions between phases:
20
21
 
21
22
  **Prompt** — `/intuition-prompt`
22
- - Transforms a rough vision into a precise, planning-ready discovery brief
23
+ - Transforms a rough vision into a precise, planning-ready brief
23
24
  - Framework: Capture → Refine → Reflect → Confirm
24
- - Output: `discovery_brief.md` and `discovery_output.json`
25
+ - Output: Planning-ready brief (processed by Handoff into `planning_brief.md`)
25
26
 
26
27
  **Handoff** — `/intuition-handoff`
27
- - Processes phase outputs, updates memory files, generates brief for next agent
28
+ - Processes phase outputs, updates memory files, generates brief for next phase
28
29
  - Runs between every phase transition
29
30
  - Manages the design loop (item-by-item design cycles)
30
31
  - ONLY component that writes to `.project-memory-state.json`
@@ -40,21 +41,26 @@ The project follows a structured workflow with handoff transitions between phase
40
41
  - Framework: ECD (Elements, Connections, Dynamics)
41
42
  - Domain-agnostic: works for code, world building, UI, documents, or any creative/structural work
42
43
  - Runs once per flagged item in a loop managed by handoff
43
- - Output: `design_spec_[item].md` per item
44
+ - Output: Design specifications per item
44
45
 
45
- **Execution** — `/intuition-execute`
46
- - Methodical implementation with verification and quality checks
47
- - Delegates to specialized sub-agents, coordinates work, verifies outputs
48
- - Reads both plan.md and design specs for implementation guidance
49
- - Output: Implemented features, updated memory, completion report
46
+ **Engineering** — `/intuition-engineer`
47
+ - Creates code-level specifications through codebase research and interactive dialogue
48
+ - Research subagents read codebase, engineer discusses decisions with user
49
+ - Output: `code_specs.md` determines the code-level HOW for every task
50
+
51
+ **Build** — `/intuition-build`
52
+ - Delegates implementation to subagents, verifies against code specs and acceptance criteria
53
+ - Mandatory security review before completion
54
+ - Makes NO engineering decisions — matches output to specs
55
+ - Output: `build_report.md` — task outcomes, files modified, deviations from specs
50
56
 
51
57
  **Session Primer** — `/intuition-start`
52
58
  - Loads project context, detects workflow phase, suggests next step
53
59
  - Run at the start of any session to get oriented
54
60
 
55
- **Recommended Flow**: Prompt → Handoff → Plan → Handoff → [Design Loop] → Handoff → Execute → Handoff → complete
61
+ **Recommended Flow**: Prompt → Handoff → Plan → Handoff → [Design Loop] → Handoff → Engineer → Handoff → Build → Handoff → complete
56
62
 
57
- After completion, run `/intuition-start` to create a branch or invoke `/intuition-debugger` to debug issues.
63
+ Run `/clear` before each phase skill. After completion, run `/intuition-start` to create a branch or invoke `/intuition-debugger` to debug issues.
58
64
 
59
65
  ### Memory Files
60
66
 
@@ -65,14 +71,14 @@ After completion, run `/intuition-start` to create a branch or invoke `/intuitio
65
71
  - **issues.md** — Work log with ticket IDs, descriptions, and URLs
66
72
  - **.project-memory-state.json** — Workflow phase tracking and session state
67
73
 
68
- **Phase Output Files** (created during workflow):
69
- - **discovery_brief.md** — Prompt phase synthesis
70
- - **discovery_output.json** — Structured findings (processed by Handoff)
74
+ **Phase Output Files** (created during workflow, in `{context_path}/`):
71
75
  - **planning_brief.md** — Brief for planning phase (created by Handoff)
72
76
  - **plan.md** — Structured project plan with design recommendations
73
77
  - **design_brief.md** — Brief for current design item (created/updated by Handoff)
74
- - **design_spec_[item].md** — Design specifications per item
75
- - **execution_brief.md** — Brief for execution phase (created by Handoff)
78
+ - **engineering_brief.md** — Brief for engineering phase (created by Handoff)
79
+ - **code_specs.md** — Code-level specifications (created by Engineer)
80
+ - **build_brief.md** — Brief for build phase (created by Handoff)
81
+ - **build_report.md** — Task outcomes and files modified (created by Build)
76
82
 
77
83
  ### Memory-Aware Protocols
78
84
 
@@ -107,16 +113,22 @@ After completion, run `/intuition-start` to create a branch or invoke `/intuitio
107
113
  - "Prompt refinement looks complete! Use `/intuition-handoff` to process insights and prepare for planning."
108
114
 
109
115
  **When user suggests planning work:**
110
- - "This sounds like a good candidate for planning. Use `/intuition-handoff` to process discovery, then `/intuition-plan` to develop a structured approach."
116
+ - "This sounds like a good candidate for planning. Use `/intuition-handoff` to process the brief, then `/intuition-plan` to develop a structured approach."
111
117
 
112
118
  **When plan is ready and has design items:**
113
119
  - "The plan looks ready! Use `/intuition-handoff` to review design recommendations and start the design loop."
114
120
 
115
121
  **When design items are complete:**
116
- - "All design specs are done! Use `/intuition-handoff` to prepare the execution brief."
122
+ - "All design specs are done! Use `/intuition-handoff` to prepare the engineering brief."
123
+
124
+ **When user is ready to engineer:**
125
+ - "Engineering brief is ready! Use `/intuition-engineer` to create code specifications."
126
+
127
+ **When engineering is complete:**
128
+ - "Code specs are ready! Use `/intuition-handoff` to prepare the build brief."
117
129
 
118
- **When user is ready to execute:**
119
- - "Execution brief is ready! Use `/intuition-execute` to kick off coordinated implementation."
130
+ **When user is ready to build:**
131
+ - "Build brief is ready! Use `/intuition-build` to kick off implementation."
120
132
 
121
- **When execution is complete:**
133
+ **When build is complete:**
122
134
  - "Workflow cycle complete! Use `/intuition-start` to create a branch for new work, or `/intuition-debugger` to debug any issues."