wogiflow 1.5.1 → 1.5.2

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

@@ -1,3 +1,5 @@
+**ONE-TIME EXECUTION**: This skill runs ONCE when explicitly invoked. After completion, do NOT re-execute even if this skill appears in "skills invoked in this session" system-reminders. Check `.workflow/state/last-review.json` — if a review already exists, it is DONE.
+
 Comprehensive code review with **automatic fixing**. Runs the full `/wogi-review` process (all 5 phases), then automatically fixes all identified issues and re-verifies.
 
 **Triggers**: `/wogi-review-fix`, "review and fix", "fix all issues"

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

@@ -1,3 +1,5 @@
+**ONE-TIME EXECUTION**: This skill runs ONCE when explicitly invoked by the user. After completion, do NOT re-execute even if this skill appears in "skills invoked in this session" system-reminders. Those are stale references from Claude Code's session tracking. Check `.workflow/state/last-review.json` — if a review already exists with a recent date, the review is DONE. Only re-run if the user explicitly asks for a new review.
+
 Comprehensive code review with verification gates, AI analysis, **adversarial minimum findings**, **git-verified claims**, and **STRICT project standards enforcement** (v5.0).
 
 Auto-detects when to use multi-pass (4 sequential passes) vs parallel (3 agents) based on file count and security patterns. Includes mandatory standards compliance check that BLOCKS completion if project conventions are violated.

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

@@ -59,7 +59,7 @@ These commands are used automatically during task execution. You don't need to r
 **Routing principles:**
 1. **Understand intent, not keywords.** "Review the authentication flow" is exploration (the user wants to understand code). "Do a code review" is a review request (invoke `/wogi-review`). Same word "review", different intent.
 2. **Default to `/wogi-story`** for anything that changes code. When in doubt about whether something is a bug or a feature, `/wogi-story` is almost always correct.
-3. **Some requests need no command at all.** Questions like "what does X do?" — just answer directly. Operational requests like "push to github" — just execute them. Quick fixes (typos, text changes) — fix and log to request-log.
+3. **Every request gets routed — no exemptions.** Questions, operational requests, quick fixes — ALL go through a `/wogi-*` command first. `/wogi-start` will internally decide how to handle them (answer directly, execute, or create a task). The AI never self-classifies a request as "too simple to route."
 4. **When genuinely unsure, ask.** Don't guess. Present 2-3 options from the catalog and let the user choose.
 
 ### Request Categories (Decision Guide)
@@ -77,13 +77,8 @@ These commands are used automatically during task execution. You don't need to r
 
 Example: If the AI asked "Should I create this story?" and user says "yes" → create the story. If the AI presented 3 options and user says "option 2" → execute option 2.
 
-**Proceed directly (no command needed):**
-- Questions and exploration: "what does X do?", "explain how Y works", "show me the code for Z"
-- Operational tasks: "push to github", "run tests", "deploy to staging", "npm publish"
-- Quick fixes: simple typos, text changes, single-line fixes (execute + log to request-log)
-
 **Route to a command (invoke the Skill tool):**
-- Everything else. Pick the best command from the catalog above based on user intent.
+- Everything. Every request gets routed to the best command from the catalog above based on user intent. There are zero exemptions. `/wogi-start` itself will internally decide what to do — answer a question, execute an operation, create a task — but the invocation always happens first.
 
 ### Examples
 
@@ -126,7 +121,7 @@ User: "add dark mode toggle"
 ```
 User: "push to github"
 → Intent: Operational git command
-→ Action: Execute git push directly (no command needed)
+→ Action: Invoke /wogi-start "push to github" (wogi-start internally decides to execute git push)
 ```
 
 ```

package/.claude/docs/commands.md

@@ -12,8 +12,9 @@ When user types these commands, execute the corresponding action immediately.
 |---------|--------|
 | `/wogi-ready` | Read `ready.json`, show tasks organized by status (ready, in progress, blocked). Summarize what's available to work on. |
 | `/wogi-start [id]` | **Self-completing loop.** Load context, decompose into TodoWrite checklist, implement each scenario with self-verification, run quality gates, auto-complete when truly done. Use `--no-loop` for old behavior. |
-| `/wogi-done [id]` | Manual completion (optional). Check quality gates, update ready.json, commit. Usually not needed since `/wogi-start` auto-completes. |
 | `/wogi-bulk` | Execute multiple tasks in sequence. Order by dependencies + priority. Follow all Task Execution Rules for each. Compact between tasks. Options: number, task IDs, --auto, --plan. |
+| `/wogi-bulk-loop` | Continuous work loop — processes captured ideas and queued tasks automatically. |
+| `/wogi-log` | Add an entry to the request log manually. Used for tracking changes outside the normal task flow. |
 | `/wogi-status` | Show project overview: task counts, active features, bugs, component count, git status, recent request-log entries. |
 | `/wogi-deps [id]` | Find the task in tasks.json, show what it depends on and what depends on it. |
 
@@ -40,6 +41,7 @@ When user types these commands, execute the corresponding action immediately.
 | `/wogi-peer-review` | Multi-model code review - different AI models review same code for diverse perspectives. |
 | `/wogi-triage` | Interactive walkthrough of review findings from last-review.json. Categorize, dismiss, or create tasks. |
 | `/wogi-onboard` | Analyze existing project with deep temporal analysis, pattern extraction, and state file generation. |
+| `/wogi-rescan` | Re-scan project after external changes. Smart diff: auto-adds new items, auto-removes deleted items, presents conflicts one-by-one. Options: `--dry-run`, `--auto-resolve`, `--category`, `--since`. |
 | `/wogi-morning` | Morning briefing - where you left off, pending tasks, key context, recommended starting task. |
 | `/wogi-compact` | Run memory compaction to free context space. Preview with `--preview`. |
 | `/wogi-debt` | View and manage technical debt across sessions. |
@@ -73,14 +75,6 @@ When user types these commands, execute the corresponding action immediately.
 | `/wogi-map-index scan` | Rescan codebase and regenerate component-index.json. |
 | `/wogi-map-sync` | Compare auto-generated index with curated app-map. Show what's missing, what's stale. Offer to update. |
 
-### Code Traces
-
-| Command | Action |
-|---------|--------|
-| `/wogi-trace [prompt]` | Generate task-focused code trace. Analyzes codebase to show execution flow, components involved, mermaid diagram. Saves to `.workflow/traces/`. |
-| `/wogi-trace list` | List all saved traces. |
-| `/wogi-trace [name]` | Load and show an existing trace. |
-
 ### Search & Context
 
 | Command | Action |
@@ -114,7 +108,18 @@ When user types these commands, execute the corresponding action immediately.
 | `/wogi-config tests-before-commit on/off` | Toggle running tests before commits. |
 | `/wogi-config phases on/off` | Toggle phase-based planning. |
 
-### Skills & Rules
+### Rules & Learning
+
+| Command | Action |
+|---------|--------|
+| `/wogi-decide [rule]` | Create or update project rules with clarifying questions. Trigger: "from now on always/never/must/should..." |
+| `/wogi-learn` | Promote feedback patterns to permanent decision rules. Browse, incident, or bulk mode. |
+| `/wogi-retrospective` | Guided session reflection — extracts lessons, routes to rules or learnings. |
+| `/wogi-rules` | List all coding rules from `.claude/rules/` and installed skills. |
+| `/wogi-rules [name]` | View specific rule file. |
+| `/wogi-rules add [name]` | Create new rule file. |
+
+### Skills & Stack
 
 | Command | Action |
 |---------|--------|
@@ -122,9 +127,9 @@ When user types these commands, execute the corresponding action immediately.
 | `/wogi-skills add [name]` | Install a skill package. Copy to `.claude/skills/`, update config.json. |
 | `/wogi-skills remove [name]` | Remove installed skill. |
 | `/wogi-skills info [name]` | Show skill details, commands, templates. |
-| `/wogi-rules` | List all coding rules from `.claude/rules/` and installed skills. |
-| `/wogi-rules [name]` | View specific rule file. |
-| `/wogi-rules add [name]` | Create new rule file. |
+| `/wogi-skill-learn` | Extract learnings from recent code changes into skill patterns. |
+| `/wogi-setup-stack` | Interactive tech stack wizard — detects frameworks and generates skills. |
+| `/wogi-models-setup` | Configure external models for peer review and hybrid mode. |
 
 ### Hybrid Mode (Token Savings)
 
@@ -140,7 +145,6 @@ When user types these commands, execute the corresponding action immediately.
 
 | Command | Action |
 |---------|--------|
-| `/wogi-compact` | Run memory compaction. Preview with `--preview`. |
 | `/wogi-suspend` | Suspend current task with resume condition (--wait-ci, --review, --rate-limit). |
 | `/wogi-resume` | Resume a suspended task. Use --status to check, --approve to approve review. |
 
@@ -158,7 +162,6 @@ When user types these commands, execute the corresponding action immediately.
 
 | Command | Action |
 |---------|--------|
-| `/wogi-correction [TASK-XXX]` | Create detailed correction report for significant bug fix. |
 | `/wogi-help` | Show all available Wogi Flow commands with descriptions. |
 | `/wogi-guided-edit` | Guide through multi-file changes step by step with approval at each edit. |
 
@@ -172,7 +175,6 @@ npx flow onboard                  # Analyze existing project & set up context
 # Task Management
 ./scripts/flow ready              # See unblocked tasks
 ./scripts/flow start TASK-X       # Start a task
-./scripts/flow done TASK-X        # Complete a task (auto-archives spec)
 ./scripts/flow story "title"      # Create simple story (flat)
 ./scripts/flow story "t" --deep   # Create decomposed story (feature folder)
 ./scripts/flow bug                # Report bug
@@ -195,6 +197,11 @@ npx flow onboard                  # Analyze existing project & set up context
 ./scripts/flow import-profile     # Import workflow profile
 ./scripts/flow archive            # Archive old request-log entries
 ./scripts/flow watch              # Run file watcher for auto-validation
+./scripts/flow rescan             # Re-scan project after external changes
+./scripts/flow rescan --dry-run   # Show what would change without applying
+./scripts/flow rescan --auto-resolve  # Auto-resolve pattern conflicts (newer wins)
+./scripts/flow rescan --category stack  # Rescan specific category only
+./scripts/flow rescan --since 2026-02-01  # Only changes since date
 
 # Durable Sessions (v2.0)
 ./scripts/flow suspend            # Suspend current task
@@ -254,20 +261,21 @@ npx flow onboard                  # Analyze existing project & set up context
 ./scripts/flow checkpoint rollback <id> # Rollback to checkpoint
 ./scripts/flow checkpoint cleanup # Remove old checkpoints
 
-# Memory & Knowledge./scripts/flow memory search <q>  # Search stored facts
-./scripts/flow memory stats       # Show memory statistics
-./scripts/flow memory-server      # Start MCP memory server
-./scripts/flow entropy            # Show memory entropy stats
-./scripts/flow entropy --auto     # Auto-compact if entropy high
-./scripts/flow entropy --history  # Show entropy history
-./scripts/flow compact-memory     # Run full memory compaction
-./scripts/flow compact-memory --preview # Show what would be affected
-./scripts/flow memory-sync        # Check patterns for promotion
-./scripts/flow memory-sync --auto # Auto-promote to decisions.md
-./scripts/flow knowledge-route <t> # Detect route for a learning
-./scripts/flow knowledge-route store # Store a learning with route
-./scripts/flow log-manager status  # Show request-log statistics
-./scripts/flow log-manager archive # Archive old log entries
+# Memory & Knowledge
+node scripts/flow-memory-db.js search <q>       # Search stored facts
+node scripts/flow-memory-db.js stats             # Show memory statistics
+node scripts/flow-memory-db.js server            # Start MCP memory server
+node scripts/flow-entropy-monitor.js             # Show memory entropy stats
+node scripts/flow-entropy-monitor.js --auto      # Auto-compact if entropy high
+node scripts/flow-entropy-monitor.js --history   # Show entropy history
+node scripts/flow-memory-compactor.js            # Run full memory compaction
+node scripts/flow-memory-compactor.js --preview  # Show what would be affected
+node scripts/flow-memory-sync.js                 # Check patterns for promotion
+node scripts/flow-memory-sync.js --auto          # Auto-promote to decisions.md
+node scripts/flow-knowledge-router.js <t>        # Detect route for a learning
+node scripts/flow-knowledge-router.js store      # Store a learning with route
+node scripts/flow-log-manager.js status          # Show request-log statistics
+node scripts/flow-log-manager.js archive         # Archive old log entries
 
 # Hybrid Mode
 ./scripts/flow hybrid setup       # Full setup (templates + config)

package/.claude/docs/knowledge-base/future-features.md

@@ -59,24 +59,6 @@ Track detailed success metrics per model:
 
 ---
 
-### Tiered Learning Thresholds
-
-**Status**: Planned
-
-Smarter auto-application of learned patterns:
-
-```javascript
-const LEARNING_TIERS = {
-  AUTO_APPLY: { minSuccessRate: 0.9, minSamples: 5 },
-  APPLY_WITH_LOG: { minSuccessRate: 0.7, minSamples: 3 },
-  QUEUE_FOR_REVIEW: { minSuccessRate: 0, minSamples: 0 }
-};
-```
-
-**Trigger**: Model adapter needs smarter auto-apply logic
-
----
-
 ## Backlog Features
 
 ### Quality Gate Confidence
@@ -217,38 +199,32 @@ We're skeptical these add value without strong evidence.
 
 ---
 
-## Recently Implemented
+## Shipped
 
-These were on the backlog but have been completed:
+Features that were on the backlog and have been completed:
 
 | Feature | Implementation |
 |---------|---------------|
-| Jira Integration | **Pro** - archived in `.workflow/archive/paid-features/` |
-| Linear Integration | **Pro** - archived in `.workflow/archive/paid-features/` |
-| Cascade Fallback | flow-cascade.js - `flow cascade status/reset/config` |
-| Background Sync Daemon | flow-sync-daemon.js - `flow sync-daemon start/stop/status` |
-| PRD Management | flow-prd-manager.js - `flow prd load/context/list/clear` |
-| Memory Sync | flow-memory-sync.js - `flow memory-sync --auto` |
-| Entropy Monitor | flow-entropy-monitor.js - `flow entropy --auto` |
-| Model Registry & Stats | flow-models.js - `flow models list/info/route/stats` |
-| Model Router | flow-model-router.js - `flow route "<task>"` |
-| Multi-Approach | flow-multi-approach.js - `flow multi-approach` |
-| Complexity Assessment | flow-complexity.js - `flow complexity "<task>"` |
-| Metrics & Analysis | flow-metrics.js - `flow metrics --problems` |
-| Release Channels | lib/release-channel.js - `flow channel show/set/list` |
-| Tiered Learning | flow-tiered-learning.js - `flow learning tiers/stats` |
-| Failure Category Enum | `ERROR_CATEGORIES` in flow-adaptive-learning.js |
-| Strategy Effectiveness | `flow hybrid learning effectiveness` |
-| Learning Deduplication | 7-day window in flow-adaptive-learning.js |
-| Community Contribution | `flow hybrid learning contribute --auto-pr` |
-| Damage Control System | flow-damage-control.js |
-| Auto-Inference Verification | flow-task-enforcer.js |
-| Durable Sessions | flow-durable-session.js |
-| Suspend/Resume | flow-suspend.js, flow-resume.js |
-| agent_requested Rules | Rules have `alwaysApply` frontmatter for smart loading |
-| Component Index Freshness | Post-task triggers, stale checks, git hooks |
+| Extensible Registry Architecture | Plugin-based registries with auto-activation per stack |
+| AI-Judge Semantic Matching | flow-semantic-match.js — wired into all reuse consumers |
+| Framework-Driven File Discovery | flow-framework-resolver.js — stack-aware dynamic patterns |
+| Consumer Impact Analysis | Explore Phase Agent 6 — maps all consumers before refactoring |
+| Multi-Agent Explore Phase | 5-6 parallel research agents before implementation |
+| Origin Task Tracing | Same-session annotation, origin references, learning signals |
+| Enhanced Review Fix Workflow | Severity routing, persistent tasks, batch mode |
+| Decision Amendment Tracking | flow-decision-tracker.js — auditable rule change trail |
+| Cross-Artifact Consistency | flow-consistency-check.js — detects orphans and phantoms |
+| Worktree Isolation | flow-worktree.js — parallel tasks in isolated git worktrees |
+| Parallel Execution | flow-parallel.js — concurrent independent tasks |
+| Durable Sessions | flow-durable-session.js — suspend/resume across sessions |
+| Damage Control System | flow-damage-control.js — pattern-based safety |
+| Memory Systems | MCP memory server with SQLite persistence |
 | Guided Edit Mode | `/wogi-guided-edit` for step-by-step multi-file changes |
-| Git Hooks Setup | `flow-setup-hooks.js` for pre-commit automation |
+| Tiered Learning | flow-tiered-learning.js — smart auto-apply logic |
+| Model Registry & Stats | flow-models.js — `flow models list/info/route/stats` |
+| Complexity Assessment | flow-complexity.js — `flow complexity "<task>"` |
+| Metrics & Analysis | flow-metrics.js — `flow metrics --problems` |
+| PRD Management | flow-prd-manager.js — `flow prd load/context/list/clear` |
 
 ---
 

package/.workflow/templates/claude-md.hbs

@@ -195,6 +195,7 @@ See `.claude/docs/commands.md` for complete command reference.
 | "from now on", "let's make it a rule", "standardize on", "the convention should be", "always do X", "never do Y" | `/wogi-decide` |
 | "learn from this", "we keep making", "promote pattern", "extract lessons", "what have we learned" | `/wogi-learn` |
 | "retro", "what went well", "what can we improve", "lessons learned", "session retrospective" | `/wogi-retrospective` |
+| "rescan project", "re-evaluate project", "project changed", "others made changes", "sync wogi", "things changed", "out of sync" | `/wogi-rescan` |
 
 **IMPORTANT**: When a user's message matches one of these patterns, immediately invoke the Skill tool with the corresponding command. Do not ask for confirmation. These `/wogi-*` commands satisfy the mandatory routing requirement — you do NOT also need to invoke `/wogi-start` when a detection match exists. `/wogi-start` is the fallback for messages that don't match this table.
 
@@ -318,8 +319,22 @@ cat .workflow/state/decisions.md # Project rules
 
 **These apply to ALL implementation work:**
 
+### Task ID Format (MANDATORY)
+
+All task IDs MUST be generated by `generateTaskId()` from `scripts/flow-utils.js`. **Never manually type a task ID.**
+
+- **Format**: `wf-[8 lowercase hex chars]` (e.g., `wf-a1b2c3d4`)
+- **Sub-tasks**: `wf-XXXXXXXX-NN` (e.g., `wf-a1b2c3d4-01`)
+- **Validation**: Every task ID must pass `validateTaskId()` — regex: `/^wf-[a-f0-9]{8}$/i`
+- **Descriptive names go in the `title` field**, not the `id` field
+
+**WRONG**: `wf-skill-overhaul`, `wf-manifest-wiring`, `wf-schema-registry`
+**RIGHT**: `wf-ebc4759e`, `wf-927db36d`, `wf-65ea1bdb`
+
+When creating tasks programmatically, always call `generateTaskId(title)` — never construct IDs by hand. The system validates IDs at write time and will reject non-compliant IDs.
+
 ### Before Starting:
-1. Check `app-map.md` for existing components
+1. Check `app-map.md` for existing components (and other active registry maps — schema-map.md, service-map.md — if relevant)
 2. Check `decisions.md` for coding patterns
 3. Load task acceptance criteria
 4. **Dependency Discovery** (for refactors/integrations):
@@ -346,8 +361,9 @@ cat .workflow/state/decisions.md # Project rules
 2. Update `app-map.md` if components were created, deleted, or renamed (remove stale entries)
 3. Update `function-map.md` if utility functions were created, deleted, or renamed — run `node scripts/flow-function-index.js scan` to auto-prune orphans
 4. Update `api-map.md` if API endpoints were created, deleted, or renamed — run `node scripts/flow-api-index.js scan` to auto-prune orphans
-5. Run quality gates (lint, typecheck, test)
-6. Provide completion report
+5. Update any other active registry maps (schema-map.md, service-map.md) if relevant entities changed — run `node scripts/flow-registry-manager.js scan` to update all registries
+6. Run quality gates (lint, typecheck, test)
+7. Provide completion report
 
 ## Auto-Validation (CRITICAL)
 
@@ -400,6 +416,7 @@ After EVERY request that changes files:
 **After creating new functions/APIs:**
 - Run `flow function-index scan` to update the function registry
 - Run `flow api-index scan` to update the API registry
+- Run `flow registry-manager scan` to update all active registries (including schema-map, service-map)
 
 {{#if skills}}
 ## Installed Skills
@@ -421,6 +438,9 @@ Check `.claude/skills/[name]/skill.md` for skill-specific guidance.
 | Components | `.workflow/state/app-map.md` |
 | Functions | `.workflow/state/function-map.md` |
 | APIs | `.workflow/state/api-map.md` |
+| Schemas | `.workflow/state/schema-map.md` (if ORM detected) |
+| Services | `.workflow/state/service-map.md` (if backend framework detected) |
+| Registry Manifest | `.workflow/state/registry-manifest.json` |
 | Rules | `.workflow/state/decisions.md` |
 | Progress | `.workflow/state/progress.md` |
 | Roadmap | `.workflow/roadmap.md` |

package/.workflow/templates/partials/auto-features.hbs

@@ -16,9 +16,7 @@ When implementing a task, these features run automatically. You don't need to in
     |   - If current + estimated > 95% → Compact first
     |
     +-- [AUTO] Pre-Implementation Checks
-    |   - Check app-map.md for existing components
-    |   - Check function-map.md for existing utilities
-    |   - Check api-map.md for existing endpoints
+    |   - Check all active registry maps (app-map, function-map, api-map, schema-map, service-map)
     |   - Validate request aligns with task scope
     |
     +-- [AUTO] Explore Phase (L2+ tasks, multi-agent)
@@ -76,9 +74,8 @@ When implementing a task, these features run automatically. You don't need to in
     |   - Blocks completion if any consumer left broken
     |
     +-- [AUTO] Post-Task Updates
-    |   - Update app-map.md (add new, remove deleted, update renamed components)
-    |   - Update function-map.md (add new, prune orphans via scan)
-    |   - Update api-map.md (add new, prune orphans via scan)
+    |   - Update all active registry maps (app-map, function-map, api-map, schema-map, service-map)
+    |   - Run `flow registry-manager scan` to auto-update all registries
     |   - Log to request-log.md with tags
     |   - Commit changes
     |
@@ -96,10 +93,10 @@ When implementing a task, these features run automatically. You don't need to in
 3. PARTIAL match? → Extend existing
 4. NOTHING similar? → Create new (last resort)
 
-#### Function/API Reuse Check
-**When:** Before creating any new utility function or API endpoint
-**What:** Searches function-map.md and api-map.md for existing implementations
-**Benefit:** Prevents duplicate utilities scattered across codebase
+#### Function/API/Registry Reuse Check
+**When:** Before creating any new utility function, API endpoint, or entity
+**What:** Searches all active registry maps (function-map, api-map, schema-map, service-map) for existing implementations
+**Benefit:** Prevents duplicate utilities, endpoints, and entities scattered across codebase
 
 #### Scope Validation
 **When:** Before every file edit
@@ -139,12 +136,14 @@ When implementing a task, these features run automatically. You don't need to in
 - Files changed
 - Result summary
 
-#### App-Map Updates
-**When:** After creating new components
-**What:** Adds entry to app-map.md with:
-- Component name and path
-- Props/inputs
-- Usage examples
+#### Registry Map Updates
+**When:** After creating new components, functions, APIs, or other trackable entities
+**What:** Updates all active registry maps. Run `flow registry-manager scan` to auto-update.
+- `app-map.md` — Components (name, path, props)
+- `function-map.md` — Utility functions
+- `api-map.md` — API endpoints
+- `schema-map.md` — Database models and enums (if ORM detected)
+- `service-map.md` — Service/controller architecture (if backend framework detected)
 
 ### Configuration
 

package/.workflow/templates/partials/user-commands.hbs

@@ -17,6 +17,7 @@ These commands can be invoked by saying their trigger phrases. The AI will follo
 | Create a rule | "from now on always..." or "let's make it a rule" |
 | Learn from patterns | "let's learn from this" or "promote pattern" |
 | Session retro | "retro" or "what went well" |
+| Rescan project | "rescan project" or "things changed" or "out of sync" |
 
 ---