wogiflow 1.8.2 → 1.8.4

package/.claude/commands/wogi-peer-review.md

@@ -1,121 +1,9 @@
 ---
 description: "Run a multi-model peer review with different AI perspectives"
 ---
-Run a multi-model peer review where different AI models review the same code.
+Run a multi-model peer review where different AI models review the same code. This command uses a **different review focus** than `/wogi-review` — it targets **improvement opportunities**, not correctness/bugs.
 
-## Step 0: Model Selection (Session Persistent)
-
-**Models are selected once per session and remembered for subsequent runs.**
-
-### Check Session State First
-
-```javascript
-const modelConfig = require('./scripts/flow-model-config');
-
-// Run migration if needed (handles old config formats)
-modelConfig.migrateOldConfig();
-
-// Check if Claude should also review
-const includeClaude = modelConfig.shouldIncludeClaude();
-
-// Check if models already selected this session
-const sessionModels = modelConfig.getSessionModels('peerReview');
-
-if (sessionModels && sessionModels.length > 0 && !args.includes('--select-models')) {
-  // Use session models - show brief note
-  const claudeNote = includeClaude ? ' + Claude' : '';
-  console.log(`Using models: ${sessionModels.join(', ')}${claudeNote}`);
-  console.log(`(Run with --select-models to change)`);
-  // Proceed with review using sessionModels
-} else {
-  // Need to select models - continue to selection flow
-}
-```
-
-### If No Models Configured
-
-If `modelConfig.getEnabledModels().length === 0`:
-
-```
-No external models configured for peer review.
-
-Run /wogi-models-setup to configure:
-- OpenAI (GPT-4o)
-- Google (Gemini)
-- Local LLM (Ollama)
-
-Or use --manual flag for manual mode.
-```
-
-Then either:
-- Auto-launch `/wogi-models-setup` wizard
-- Or use `--manual` mode if user prefers
-
-### Model Selection Dialog
-
-Show selection dialog when:
-- First run of session (no session models set)
-- User passes `--select-models` flag
-
-```javascript
-{
-  question: "Select models for peer review (multiple allowed):",
-  header: "Models",
-  multiSelect: true,
-  options: [
-    // Claude option (when includeClaude is enabled in config)
-    { label: "Claude (current session)", description: "Reviews using current conversation context" },
-    // Dynamically populated from configured models
-    { label: "openai:gpt-4o", description: "Best quality reasoning" },
-    { label: "openai:gpt-4o-mini", description: "Faster, cheaper" },
-    { label: "google:gemini-2.0-flash", description: "Fast, good at code" },
-    { label: "local:qwen2.5-coder", description: "Free, runs locally" }
-    // ... other configured models
-  ]
-}
-```
-
-**Show Claude option when:**
-- `modelConfig.shouldIncludeClaude()` returns `true`
-- Claude is shown first (recommended) as it has full context
-
-**Show external models that:**
-1. Are configured in `models.providers`
-2. Have `enabled: true`
-3. Have API key set (check `process.env[apiKeyEnv]`) or are local
-
-### After Selection
-
-Save the selection to session state:
-```javascript
-modelConfig.setSessionModels('peerReview', selectedModels);
-```
-
-This persists until `/wogi-session-end` is called, which clears the selection.
-
-Then proceed with the review using selected models.
-
-## How It Works
-
-### When `includeClaude` is enabled (recommended):
-1. **Claude reviews first** using the same improvement-focused prompt as external models
-2. **External model(s)** review the same changes via API
-3. **All findings compared** (Claude + external models)
-4. **Claude synthesizes** all perspectives and responds to disagreements
-
-### When `includeClaude` is disabled:
-1. **External model(s)** review the changes via API
-2. **Findings are compared** across external models
-3. **Claude synthesizes** findings and responds to peer feedback:
-   - Defends decisions with context
-   - OR acknowledges valid alternatives
-
-**Why include Claude?**
-- Provides additional perspective alongside external models
-- Has full conversation context (knows why certain decisions were made)
-- Catches things external models might miss due to context limitations
-
-## Key Difference from `/wogi-review`
+## Relationship to /wogi-review
 
 | `/wogi-review` | `/wogi-peer-review` |
 |----------------|---------------------|
@@ -124,14 +12,9 @@ Then proceed with the review using selected models.
 | Security vulnerabilities | Alternative implementations |
 | Architecture conflicts | Pattern suggestions |
 | Verification-focused | Improvement-focused |
+| Single model (multi-agent) | Multiple external models + optional Claude |
 
-## What Peer Review Surfaces
-
-1. **Optimization opportunities** - "This works, but could be faster/cleaner"
-2. **Alternative approaches** - "Consider doing X instead of Y"
-3. **Cross-model disagreements** - Where different models see things differently
-4. **Pattern suggestions** - "Other codebases typically do this as..."
-5. **Missed edge cases** - Fresh eyes catch what familiarity misses
+This command does NOT run `/wogi-review`'s 5-phase process. Instead it runs its own flow: collect changes, generate improvement-focused prompt, send to multiple models, compare and synthesize results.
 
 ## Usage
 
@@ -143,81 +26,17 @@ Then proceed with the review using selected models.
 /wogi-peer-review --manual           # Manual mode (no API keys needed)
 ```
 
-## Provider Configuration
-
-### Recommended: Use `/wogi-models-setup`
-
-The easiest way to configure models is the setup wizard:
-```
-/wogi-models-setup
-```
-
-This creates a unified configuration used by both peer review and hybrid mode.
-
-### Config Location
-
-Models are configured in `.workflow/config.json` under `models`:
-
-```json
-{
-  "models": {
-    "providers": {
-      "openai": {
-        "apiKeyEnv": "OPENAI_API_KEY",
-        "enabled": true,
-        "models": ["gpt-4o", "gpt-4o-mini"]
-      },
-      "google": {
-        "apiKeyEnv": "GOOGLE_API_KEY",
-        "enabled": true,
-        "models": ["gemini-2.0-flash"]
-      },
-      "local": {
-        "endpoint": "http://localhost:11434",
-        "provider": "ollama",
-        "enabled": true,
-        "models": ["qwen2.5-coder"]
-      }
-    },
-    "defaults": {
-      "peerReview": ["openai:gpt-4o", "google:gemini-2.0-flash"],
-      "includeClaude": true
-    }
-  }
-}
-```
+## Step 0: Model Selection (Session Persistent)
 
-API keys are stored in `.env` (not in config):
-```
-OPENAI_API_KEY=sk-proj-...
-GOOGLE_API_KEY=AIza...
-```
+Models are selected once per session and remembered for subsequent runs.
 
-### Legacy Config (Auto-Migrated)
-
-Old format configs are automatically migrated on first use:
-```json
-// Old format (still supported, auto-migrates)
-"peerReview": {
-  "apiKeys": {
-    "openai": "${OPENAI_API_KEY}"
-  },
-  "models": ["openai:gpt-4o"]
-}
-```
+1. Check `modelConfig.getSessionModels('peerReview')` for existing selection
+2. If models already selected AND no `--select-models` flag, reuse them
+3. If no models configured (`modelConfig.getEnabledModels().length === 0`), prompt to run `/wogi-models-setup` or use `--manual`
+4. Show multi-select dialog with configured models + Claude option (when `modelConfig.shouldIncludeClaude()`)
+5. Save selection: `modelConfig.setSessionModels('peerReview', selectedModels)`
 
-### Manual Mode
-
-For manual review (no API keys needed):
-```
-/wogi-peer-review --manual
-```
-
-When manual:
-1. Outputs the review prompt
-2. User runs in Cursor/other tool
-3. User pastes results back
-4. Claude synthesizes
+**Config**: Models configured in `.workflow/config.json` under `models.providers`. API keys in `.env`.
 
 ## Review Flow
 
@@ -228,55 +47,21 @@ When manual:
 │  1. Collect code changes (git diff or specified files)   │
 │  2. Generate improvement-focused prompt                  │
 │  3. If includeClaude enabled:                            │
-│     • Launch Claude review (Task agent, Explore type)   │
-│     • Claude reviews using same prompt as external       │
+│     - Launch Claude review (Task agent, Explore type)    │
 │  4. External model(s) review via API                     │
 │  5. Collect all results                                  │
 │  6. Compare findings:                                    │
-│     • All agree → Strong suggestion                      │
-│     • Partial agree → Present perspectives               │
-│     • Disagree → Surface disagreement                    │
-│  7. Claude synthesizes and responds to feedback:         │
-│     • "I have more context, here's why X is better..."   │
-│     • "Valid point, Y would be an improvement..."        │
+│     - All agree → Strong suggestion                      │
+│     - Partial agree → Present perspectives               │
+│     - Disagree → Surface disagreement                    │
+│  7. Claude synthesizes and responds to feedback           │
 │  8. Output final synthesis                               │
 └─────────────────────────────────────────────────────────┘
 ```
 
-### Claude Review Implementation
-
-When `includeClaude` is enabled, launch a Task agent to perform Claude's review:
-
-```javascript
-// In wogi-peer-review execution
-const modelConfig = require('./scripts/flow-model-config');
-
-if (modelConfig.shouldIncludeClaude()) {
-  // Launch Task agent with subagent_type=Explore
-  // Use the same improvement-focused prompt as external models
-  // The agent reviews the code and returns findings
-  // Add Claude's results to the comparison alongside external model results
-}
-```
-
-**Task agent prompt for Claude review:**
-```
-Review this code for IMPROVEMENT OPPORTUNITIES (not bugs):
-
-1. Optimization: Can this be faster/more efficient?
-2. Alternatives: Are there better approaches?
-3. Patterns: Does this follow best practices?
-4. Readability: Could this be clearer/simpler?
-5. Extensibility: Will this be easy to extend?
-
-[code changes]
-
-Return structured findings with specific suggestions.
-```
-
 ## Review Prompt Template
 
-The peer review focuses on improvements, not correctness:
+The peer review prompt focuses on improvements, not correctness:
 
 ```
 Review this code for IMPROVEMENT OPPORTUNITIES, not bugs:
@@ -287,108 +72,82 @@ Review this code for IMPROVEMENT OPPORTUNITIES, not bugs:
 4. **Readability**: Could this be clearer/simpler?
 5. **Extensibility**: Will this be easy to extend?
 
-Code:
-[code changes]
-
-Respond with:
-- Specific improvement suggestions
-- Alternative approaches considered
-- Trade-off analysis for any changes
+Respond with: specific suggestions, alternative approaches, trade-off analysis.
 ```
 
 ## Output
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-🔍 Peer Review Results
+Peer Review Results
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Reviewers: Claude, GPT-4o, Gemini 2.0 Flash
 
-✅ Agreement (3/3 models):
-   • Consider using early return for readability
-   • Extract repeated logic to helper function
+Agreement (3/3 models):
+   - Consider using early return for readability
+   - Extract repeated logic to helper function
 
-⚖️ Partial Agreement (2/3 models):
-   • Claude + Gemini: Add input validation at boundary
-   • GPT-4o: Not necessary for internal function
+Partial Agreement (2/3 models):
+   - Claude + Gemini: Add input validation at boundary
+   - GPT-4o: Not necessary for internal function
 
-⚖️ Disagreement:
-   • Claude: Prefer inline styling for this case
-   • GPT-4o: Recommend extracting to CSS module
-   • Gemini: No strong opinion
+Disagreement:
+   - Claude: Prefer inline styling for this case
+   - GPT-4o: Recommend extracting to CSS module
+   - Gemini: No strong opinion
    → Resolution: Context-dependent, current approach is valid
 
-💡 Unique Insights:
-   • [Claude] Current architecture handles edge case X well
-   • [GPT-4o] Consider memoization for expensive computation
-   • [Gemini] Similar pattern used in popular library Y
+Unique Insights:
+   - [Claude] Current architecture handles edge case X well
+   - [GPT-4o] Consider memoization for expensive computation
+   - [Gemini] Similar pattern used in popular library Y
 
-📊 Summary:
+Summary:
    Reviewers: 3 (Claude + 2 external)
    4 actionable improvements identified
    1 disagreement resolved
    Code quality: Good, with minor optimization opportunities
 ```
 
-## When to Use
-
-- Before merging significant changes
-- For security-sensitive code
-- When you want high confidence
-- For learning different perspectives
-- When stuck on architecture decisions
-
-## Phase: Post-Review Actions
+## Post-Review: Task Creation
 
-After peer review completes, optionally create tasks from actionable improvements.
-
-### Store & Create Tasks
-
-Unlike `/wogi-review` (which finds bugs), peer review finds **improvement opportunities**. These are optional enhancements, not required fixes.
+Unlike `/wogi-review` (which finds bugs and creates fix tasks), peer review finds **improvement opportunities**. These are optional enhancements, not required fixes.
 
 **Task creation rules:**
 - Strong agreements (2+ models) → Create task if user approves
 - Single-model suggestions → Note in tech-debt.json for future
 - Disagreements → Document in review report, no task
 
-**Present options:**
-```
-═══════════════════════════════════════
-ACTIONABLE IMPROVEMENTS
-═══════════════════════════════════════
-3 improvements with strong agreement:
-• Extract repeated logic to helper (readability)
-• Add memoization for expensive computation (performance)
-• Use early return pattern (readability)
-
-Options:
-[1] Create tasks - Add as improvement tasks (P3)
-[2] Add to tech-debt - Track for future
-[3] Skip - Just log the review
-```
+Present options: [1] Create tasks (P3), [2] Add to tech-debt, [3] Skip (just log).
 
-### Learning Loop
+## Post-Review: Learning Loop
 
 For recurring suggestions across reviews:
+1. Same improvement suggested 3+ times → Consider adding to `decisions.md`
+2. Pattern disagreement resolved consistently → Document the resolution
 
-1. If same improvement suggested 3+ times → Consider adding to decisions.md
-2. If pattern disagreement resolved consistently → Document the resolution
+## Manual Mode
 
-Example:
-```
-Pattern "prefer-early-return" suggested 4 times across reviews.
-Add to coding standards? [Y/n]
-```
+For manual review (no API keys needed): `/wogi-peer-review --manual`
+
+1. Outputs the review prompt with code changes
+2. User runs in Cursor/other AI tool
+3. User pastes results back
+4. Claude synthesizes all perspectives
 
 ## Options
 
-- `--select-models` - Force model re-selection (overrides session selection)
-- `--manual` - Manual mode (copy prompt to another AI, paste response back)
-- `--provider <name>` - Override configured provider
-- `--model <name>` - Specify secondary model
-- `--files <glob>` - Review specific files
-- `--task <id>` - Review task changes
-- `--json` - Output JSON for automation
-- `--verbose` - Show full model responses
-- `--create-tasks` - Auto-create tasks for strong agreements
+| Flag | Description |
+|------|-------------|
+| `--select-models` | Force model re-selection (overrides session selection) |
+| `--manual` | Manual mode (copy prompt to another AI, paste response back) |
+| `--provider <name>` | Override configured provider |
+| `--model <name>` | Specify secondary model |
+| `--files <glob>` | Review specific files |
+| `--task <id>` | Review task changes |
+| `--json` | Output JSON for automation |
+| `--verbose` | Show full model responses |
+| `--create-tasks` | Auto-create tasks for strong agreements |
+
+ARGUMENTS: {args}

package/.claude/commands/wogi-register.md

@@ -52,7 +52,18 @@ If MCP inspection yields few or no results:
 
 ### Step 3: Build Plugin Entry
 
-From the discovered information, construct:
+From the discovered information, construct each capability entry. For each capability, classify its **mode** and **flowPhases**:
+
+**Mode Classification Guide:**
+| Mode | When to use | Examples |
+|------|-------------|----------|
+| `standalone` | Action works independently, no task context needed | Screenshots, status checks, lookups, notifications, sending messages |
+| `flow-integrated` | Action is part of a development workflow phase | Running tests (validating), design checks (exploring), deploys (completing), linting (validating) |
+| `trigger` | Action feeds events INTO WogiFlow routing | Error alerts → `/wogi-bug`, deployment failures → `/wogi-story`, PR comments → `/wogi-review` |
+
+**flowPhases** (only for `flow-integrated` mode): `exploring`, `coding`, `validating`, `completing`
+
+**outputFormat**: `text` (default), `json` (structured data), `artifact` (file/image output)
 
 ```json
 {
@@ -66,7 +77,10 @@ From the discovered information, construct:
       "description": "What this action does",
       "triggerPhrases": ["send to X", "push to X"],
       "mcpTool": "mcp__server__tool_name or null",
-      "requiresTask": false
+      "requiresTask": false,
+      "mode": "standalone|flow-integrated|trigger",
+      "flowPhases": ["validating"],
+      "outputFormat": "text"
     }
   ],
   "metadata": {
@@ -91,10 +105,11 @@ Source: MCP tools | Web search | Manual
 
 Capabilities discovered (N):
   1. <action>: <description>
-     Triggers: "phrase 1", "phrase 2"
+     Mode: standalone | Triggers: "phrase 1", "phrase 2"
      MCP Tool: mcp__server__tool
 
   2. <action>: <description>
+     Mode: flow-integrated | Phases: validating, completing
      Triggers: "phrase 3"
 
 Trigger phrases (top-level):

package/.claude/commands/wogi-research.md

@@ -198,7 +198,6 @@ In `.workflow/config.json`:
   "research": {
     "enabled": true,
     "defaultDepth": "standard",
-    "strictMode": true,
     "autoTrigger": true,
     "maxTokensPerDepth": {
       "quick": 5000,
@@ -265,7 +264,7 @@ This will:
 
 ## Integration with Hooks
 
-When `research.strictMode` is enabled and `research.autoTrigger` is true:
+When `research.requireCitations` is enabled and `research.autoTrigger` is true:
 - Capability/feasibility questions automatically trigger research
 - Claims without citations are flagged
 - Negative claims require exhaustive search evidence