Data-Wise · Data-Wise · Feb 7, 2026 · Feb 6, 2026 · Feb 6, 2026 · Feb 6, 2026
diff --git a/.STATUS b/.STATUS
@@ -4,15 +4,15 @@ release_date: 2026-02-06
 milestone: v2.16.0 - Teaching Ecosystem + Branch Protection
 last_session: 2026-02-07
 
-progress: Teaching ecosystem merged, branch protection pending
+progress: Teaching ecosystem merged, branch protection complete, ready for PR
 
 📊 Branch Status:
 
 | Branch | Head    | Base | Status                          |
 |--------|---------|------|---------------------------------|
 | main   | bbd6ac7 | —    | Production (v2.15.0 released)   |
 | dev    | 0f94d9d | main | Teaching ecosystem merged (PR #57) |
-| feature/branch-protection  | 2087a98 | dev | ORCHESTRATE ready               |
+| feature/branch-protection  | —       | dev | Implementation complete, PR ready |
 
 ✅ v2.15.0 — Brainstorm v2.5.0: Spec Simplification + Smart Questions (2026-02-06):
 - Brainstorm spec simplification: 1,919 → 312 lines (84% reduction)
@@ -38,16 +38,16 @@ progress: Teaching ecosystem merged, branch protection pending
 **feature/teaching-ecosystem**: ✅ MERGED (PR #57, 2026-02-07)
 - Config normalizer, break validation fix, 8 new tests, Teaching docs tab
 
-**feature/branch-protection** (9 steps):
-1. Hook script (~/.claude/hooks/branch-guard.sh)
-2. Hook config (settings.json PreToolUse registration)
-3. Dry-run test
-4. Enable enforcement
-5. Per-project config (.claude/branch-guard.json)
-6. Craft command enhancements (check, do, worktree, status)
-7. New commands (unprotect, protect)
-8. Tests (17 hook + 4 integration)
-9. Documentation
+**feature/branch-protection** (9/9 complete ✅):
+1. ✅ Hook script (scripts/branch-guard.sh — moved into repo)
+2. ✅ Hook config (settings.json PreToolUse registration)
+3. ✅ Dry-run test
+4. ✅ Enable enforcement
+5. ✅ Per-project config (.claude/branch-guard.json)
+6. ✅ Craft command enhancements (check, do, worktree, status)
+7. ✅ New commands (unprotect, protect)
+8. ✅ Tests (49 unit + 31 e2e + 6 integration = 86 total)
+9. ✅ Documentation + standalone installer + Homebrew formula
 
 ## Future (v2.17.0+)
 

diff --git a/.claude-plugin/plugin.json b/.claude-plugin/plugin.json
@@ -1,7 +1,7 @@
 {
   "name": "craft",
   "version": "2.15.0",
-  "description": "Full-stack developer toolkit with integrated workflow automation - 106 commands (94 craft + 12 workflow), 8 agents, 21 skills with standardized dry-run preview mode (29 with --dry-run support). Code, git, site, docs, testing, architecture, CI, distribution, teaching workflows, planning, and ADHD-friendly features including brainstorming, task management, and spec capture",
+  "description": "Full-stack developer toolkit with integrated workflow automation - 108 commands (96 craft + 12 workflow), 8 agents, 21 skills with standardized dry-run preview mode (29 with --dry-run support). Code, git, site, docs, testing, architecture, CI, distribution, teaching workflows, planning, and ADHD-friendly features including brainstorming, task management, and spec capture",
   "author": {
     "name": "Data-Wise",
     "email": "dt@stat-wise.com"

diff --git a/.gitignore b/.gitignore
@@ -21,8 +21,9 @@ env/
 .DS_Store
 Thumbs.db
 
-# Test reports
+# Test reports and logs
 tests/craft_test_report.md
+tests/cli/logs/
 .coverage
 
 # Brainstorm documents (working drafts)

diff --git a/CLAUDE.md b/CLAUDE.md
@@ -2,10 +2,10 @@
 
 > **TL;DR**: Use `/craft:do <task>` for smart routing, `/craft:check` before commits, `/craft:git:worktree` for feature branches. **Always start work from `dev` branch** - never commit to `main` directly.
 
-**106 commands** · **21 skills** · **8 agents** · **23 specs** · [Documentation](https://data-wise.github.io/craft/) · [GitHub](https://github.com/Data-Wise/craft)
+**108 commands** · **21 skills** · **8 agents** · **23 specs** · [Documentation](https://data-wise.github.io/craft/) · [GitHub](https://github.com/Data-Wise/craft)
 
 **Current Version:** v2.15.0 | **Latest Release:** v2.15.0 (2026-02-06)
-**Documentation Status:** 99% complete | **Tests:** 1294 passing (176 claude-md + 998 core + 74 formatting + 38 brainstorm-context + 8 teaching-normalization)
+**Documentation Status:** 99% complete | **Tests:** 1380 passing (176 claude-md + 998 core + 74 formatting + 38 brainstorm-context + 8 teaching-normalization + 86 branch-guard)
 
 ## Git Workflow
 
@@ -34,6 +34,17 @@ feature/* (worktrees) ← All implementation work
 - **Never** write feature code on `dev`
 - **Always** verify branch: `git branch --show-current`
 
+### Branch Protection (Enforced by Hook)
+
+| Branch | Code Files | .md Files | Git Operations |
+|--------|-----------|-----------|----------------|
+| `main` | BLOCKED | BLOCKED | Commit/push BLOCKED |
+| `dev` | New: BLOCKED, Existing: allowed | ALLOWED | Commit/push allowed |
+| `feature/*` | ALLOWED | ALLOWED | All allowed |
+
+Override: `/craft:git:unprotect` (session-scoped, auto-expires)
+Config: `.claude/branch-guard.json` (per-project, optional)
+
 ## Quick Commands
 
 | Task              | Shell                                      | Craft                            |
@@ -91,7 +102,7 @@ Auto-selection: debug (errors), optimize (performance), release (deploy), else d
 ```text
 craft/
 ├── .claude-plugin/     # Plugin manifest, hooks, validators
-├── commands/           # 106 commands (arch, ci, code, docs, git, site, test, workflow)
+├── commands/           # 108 commands (arch, ci, code, docs, git, site, test, workflow)
 ├── skills/             # 21 specialized skills
 ├── agents/             # 8 agents
 ├── scripts/            # 30+ utility scripts (dependency management, converters, installers)
@@ -107,43 +118,25 @@ craft/
 
 ## Recent Major Features
 
-### v2.15.0 - Brainstorm v2.5.0: Spec Simplification + Smart Questions (2026-02-06) ✅
+### v2.16.0 - Branch Protection Hooks (2026-02-06) ✅
 
-**Part 1: Spec Simplification** — brainstorm.md reduced from 1,919 → 312 lines (84% reduction). Extracted to:
+**New Hook:** `~/.claude/hooks/branch-guard.sh` (~290 lines) — PreToolUse hook enforcing branch protection (main=block-all, dev=block-new-code, feature=unrestricted). Per-project config via `.claude/branch-guard.json`.
 
-- `docs/specs/SPEC-brainstorm-question-bank.md` — Full question bank + project-type extensions
-- `docs/tutorials/TUTORIAL-brainstorm-power-user.md` — Detailed examples + advanced patterns
-- `docs/reference/REFCARD-BRAINSTORM.md` — Flowcharts + quick reference card
+**New Commands:** `/craft:git:unprotect` + `/craft:git:protect` (session-scoped bypass)
+**Enhanced:** `/craft:check` (branch context), `/craft:do` (branch-aware routing), `/craft:git:worktree` (main block), `/craft:git:status` (guard indicator)
+**Tests:** 42 unit + 6 integration, all passing. **Files Changed:** 12 (+2,200 lines)
 
-**Part 2: Context-Aware Smart Questions** — New Step 1.7 context scan before presenting questions:
-
-- `utils/brainstorm_context.py` (~280 lines) — Scans .STATUS, specs, git log, CLAUDE.md
-- Project-type question extensions: 12 new questions (2 per type: R, Python, Node, Quarto, Plugin, Teaching)
-- Dynamic questions: matching specs, prior brainstorms, failing tests
-- Pre-fills answers from project state (version, current task)
+---
 
-**Tests:** 38 new tests (test_brainstorm_context.py), all passing
+### v2.15.0 - Brainstorm v2.5.0: Spec Simplification + Smart Questions (2026-02-06) ✅
 
-**Files Changed:** 8 (+1,500/-1,600)
+Brainstorm.md reduced 1,919 → 312 lines (84% reduction). New `utils/brainstorm_context.py` (~280 lines) scans .STATUS, specs, git log for context-aware smart questions. 12 project-type extensions, dynamic questions from project state. **Tests:** 38 new. **Files Changed:** 8 (+1,500/-1,600)
 
 ---
 
 ### v2.14.0 - Unified Formatting Library (2026-02-05) ✅
 
-**Branch:** `feature/styled-output`
-
-**New Library:** `scripts/formatting.sh` (~180 lines) — shared bash formatting library providing box-drawing (double/single line), `FMT_` prefixed color constants, ANSI-aware padding, table formatting, and source guard. All boxes standardized to 63 visible characters.
-
-**API:** `box_header`, `box_single`, `box_row`, `box_separator`, `box_footer`, `box_empty_row`, `box_table`, `fmt_set_width`, `fmt_divider`, `_fmt_strip_ansi`.
-
-**Migrations:**
-
-- 8 box-drawing scripts migrated (install.sh, migrate-from-workflow.sh, convert-cast.sh, health-check.sh, consent-prompt.sh, dependency-installer.sh, dependency-manager.sh)
-- 15 color-only scripts migrated (validate-counts, pre-release-check, batch-convert, repair-tools, 3 installers, tool-detector, version-check, sync-version, verify-phase1/2, install-hooks, test-fix-flag, pre-commit-markdownlint)
-
-**Tests:** 74 new tests (28 unit + 30 integration + 16 edge cases)
-
-**Files Changed:** 24 (+1,100/-300)
+`scripts/formatting.sh` (~180 lines) — shared box-drawing, `FMT_` color constants, ANSI-aware padding. 23 scripts migrated (8 box-drawing + 15 color-only). All boxes standardized to 63 visible characters. **Tests:** 74 new. **Files Changed:** 24 (+1,100/-300)
 
 ---
 
@@ -527,6 +520,12 @@ See `docs/specs/` for detailed specifications (23 total). See `docs/VERSION-HIST
 | `scripts/pre-release-check.sh`                    | Pre-release validation (version, counts, clean tree)    |
 | `scripts/docs-lint-emoji.sh`                      | Standalone CRAFT-001 check for pre-commit hook          |
 | `.prettierignore`                                 | Prevents prettier from breaking emoji-attribute spacing |
+| `.claude/branch-guard.json`                       | Per-project branch protection config (optional)         |
+| `commands/git/unprotect.md`                       | Session-scoped bypass for branch protection             |
+| `commands/git/protect.md`                         | Re-enable branch protection                             |
+| `tests/test_branch_guard.sh`                      | Branch guard hook unit tests (49 tests)                 |
+| `tests/test_branch_guard_e2e.sh`                  | Branch guard e2e tests (31 tests)                       |
+| `tests/test_integration_branch_guard.py`          | Branch guard integration tests (6 tests)                |
 
 ## Test Suite
 
@@ -543,17 +542,20 @@ See `docs/specs/` for detailed specifications (23 total). See `docs/VERSION-HIST
 | `tests/test_claude_md_v3.py`                       | 51       | 100%     | v3 sync/optimizer (v2.12.0)  |
 | `tests/test_claude_md_audit.py`                    | 11       | 100%     | Audit module (v2.10.0)       |
 | `tests/test_brainstorm_context.py`                 | 38       | 100%     | Context scanner (v2.15.0)    |
+| `tests/test_branch_guard.sh`                       | 49       | 100%     | Branch guard hook (v2.16.0)  |
 | **Integration & E2E Tests**                        |          |          |                              |
 | `tests/test_command_enhancements_e2e.py`           | 93       | 100%     | Command enhancements (v2.9.0)|
 | `tests/test_integration_brainstorm_phase1.py`      | 24       | 100%     | Question control integration |
 | `tests/test_integration_dependency_system.py`      | 9        | 100%     | Dependency workflow          |
 | `tests/test_integration_orchestrator_workflows.py` | 13       | 100%     | Task routing & scoring       |
 | `tests/test_integration_claude_md_v3.py`           | 9        | 100%     | v3 sync/optimizer integ.     |
 | `tests/test_integration_teaching_workflow.py`      | 16       | 100%     | Teaching mode + normalization (2 skipped) |
+| `tests/test_integration_branch_guard.py`           | 6        | 100%     | Branch guard integration     |
+| `tests/test_branch_guard_e2e.sh`                   | 31       | 100%     | Branch guard e2e (v2.16.0)   |
 | **System Tests**                                   |          |          |                              |
 | `tests/test_dependency_management.sh`              | 79       | 100%     | Dependency system            |
 | `tests/test_formatting.sh`                         | 74       | 100%     | Formatting library (v2.14.0) |
-| **Total**                                          | **1294** | **~90%** | **All systems**              |
+| **Total**                                          | **1380** | **~90%** | **All systems**              |
 
 ## Troubleshooting
 
@@ -575,7 +577,7 @@ See `docs/specs/` for detailed specifications (23 total). See `docs/VERSION-HIST
 ## Links
 
 - [Documentation Site](https://data-wise.github.io/craft/) — Full guides and references
-- [Commands Reference](https://data-wise.github.io/craft/commands/) — All 106 commands
+- [Commands Reference](https://data-wise.github.io/craft/commands/) — All 108 commands
 - [Architecture Guide](https://data-wise.github.io/craft/architecture/) — How Craft works
 - [Specifications](docs/specs/) — Implementation specs (23 total)
 - [Version History](docs/VERSION-HISTORY.md) — Complete release timeline (NEW)

diff --git a/README.md b/README.md
@@ -6,7 +6,7 @@
 [![Documentation](https://img.shields.io/badge/docs-99%25%20complete-brightgreen.svg)](https://data-wise.github.io/craft/)
 
 > **v2.15.0 - Brainstorm v2.5.0: Spec Simplification + Smart Questions** 🚀
-> **106 commands** | **21 skills** | **8 agents** | **1294 tests passing**
+> **108 commands** | **21 skills** | **8 agents** | **1380 tests passing**
 > Brainstorm spec reduced 84% (1,919 → 312 lines) with context-aware smart questions.
 
 A comprehensive production-ready toolkit for Claude Code featuring smart orchestration, ADHD-friendly workflows, multi-agent coordination, and complete documentation coverage.
@@ -135,7 +135,7 @@ Craft is a pure plugin that uses built-in Claude Code capabilities. No external
 - [Quick Start](https://data-wise.github.io/craft/QUICK-START/) (30 seconds)
 - [ADHD Guide](https://data-wise.github.io/craft/ADHD-QUICK-START/) (neurodivergent-friendly)
 - [Visual Workflows](https://data-wise.github.io/craft/workflows/) (10 GIF demonstrations)
-- [Command Reference](https://data-wise.github.io/craft/REFCARD/) (all 106 commands)
+- [Command Reference](https://data-wise.github.io/craft/REFCARD/) (all 108 commands)
 - [Skills & Agents](https://data-wise.github.io/craft/guide/skills-agents/) (21 skills, 8 agents)
 - [Claude Code 2.1 Integration](https://data-wise.github.io/craft/guide/claude-code-2.1-integration/) (comprehensive guide with 9 diagrams)
 - [Complexity Scoring Algorithm](https://data-wise.github.io/craft/guide/complexity-scoring-algorithm/) (complete technical documentation with 8 diagrams)

diff --git a/commands/check.md b/commands/check.md
@@ -53,6 +53,7 @@ Preview which checks will be performed without actually executing them:
 │   - Build tool: uv                                            │
 │   - Config: pyproject.toml                                    │
 │   - Worktree: No (main repo)                                  │
+│   - Guard: Active (block-new-code)                            │
 │   - Git status: Clean working tree                            │
 │                                                               │
 │ ✓ Validation Plan (5 checks):                                 │
@@ -149,6 +150,7 @@ Pre-flight Check Plan:
   Project: <project-name> (<project-type>)
   Mode: <mode>
   Branch: <current-branch>
+  Guard: <protection-status>
   Context: <for-value or "general">
 
   Checks to run:
@@ -286,6 +288,7 @@ fi
 │ 🌳 Worktree: ~/.git-worktrees/scribe/feat-hud       │
 │    Main: ~/projects/dev-tools/scribe                │
 │    Branch: feat/mission-control-hud                 │
+│    Guard: None (feature branches unrestricted)      │
 ├─────────────────────────────────────────────────────┤
 │ ✓ Lint         0 issues                             │
 │ ...                                                 │

diff --git a/commands/do.md b/commands/do.md
@@ -107,12 +107,58 @@ Preview which commands will be executed without actually running them:
 
 **Note**: Dry-run shows routing decision based on complexity score. Agent delegation triggers for medium (4-7) and complex (8-10) tasks.
 
+## Branch-Aware Routing (NEW in v2.16.0)
+
+Before routing, check branch protection status. When on `dev` or `main` and the task involves code changes:
+
+### On `dev` (block-new-code)
+
+If task involves creating new files or writing code:
+
+```json
+{
+  "questions": [{
+    "question": "You're on dev (protected). How should I handle this code task?",
+    "header": "Branch",
+    "multiSelect": false,
+    "options": [
+      {
+        "label": "Create worktree (Recommended)",
+        "description": "Auto-create feature branch and worktree from task description."
+      },
+      {
+        "label": "Write spec only",
+        "description": "Create a spec file on dev (allowed) without code changes."
+      },
+      {
+        "label": "Analyze only",
+        "description": "Read and analyze code without making edits."
+      }
+    ]
+  }]
+}
+```
+
+### On `main` (block-all)
+
+```
+Cannot route code tasks on main. All changes go through PRs.
+
+Switch to dev: git checkout dev
+Then retry: /craft:do <task>
+```
+
+### On `feature/*`
+
+Route directly without branch intervention — no restrictions.
+
 ## How It Works
 
-1. **Check Spec** - Look for existing spec matching task (NEW in v1.1.0)
-2. **Analyze** - Parse task description for intent and category
-3. **Score Complexity** - Calculate 0-10 score based on 5 factors (NEW in v1.23.0)
-4. **Route Decision** - Choose execution strategy:
+1. **Check Branch** - Verify branch protection status (NEW in v2.16.0)
+2. **Check Spec** - Look for existing spec matching task (NEW in v1.1.0)
+3. **Analyze** - Parse task description for intent and category
+4. **Score Complexity** - Calculate 0-10 score based on 5 factors (NEW in v1.23.0)
+5. **Route Decision** - Choose execution strategy:
    - **Score 0-3**: Route to craft commands (traditional)
    - **Score 4-7**: Delegate to specialized agent (NEW)
    - **Score 8-10**: Delegate to orchestrator-v2 (NEW)

diff --git a/commands/docs/claude-md/init.md b/commands/docs/claude-md/init.md
@@ -69,7 +69,7 @@ Analyzing project structure...
 
 Detected indicators:
   ✓ .claude-plugin/plugin.json exists
-  ✓ commands/ directory (106 commands)
+  ✓ commands/ directory (108 commands)
   ✓ skills/ directory (21 skills)
   ✓ agents/ directory (8 agents)
 
@@ -114,8 +114,8 @@ Generated CLAUDE.md Preview
 
 > **TL;DR**: Development workflow orchestration plugin
 
-**106 commands** · **21 skills** · **8 agents**
-**Version:** v2.15.0 | **Tests:** 1294 passing
+**108 commands** · **21 skills** · **8 agents**
+**Version:** v2.15.0 | **Tests:** 1380 passing
 
 ## Git Workflow
 [...]