orbital-command 0.1.0

package/templates/presets/minimal.json

@@ -0,0 +1,23 @@
+{
+  "name": "Minimal",
+  "version": 1,
+  "branchingMode": "trunk",
+  "lists": [
+    { "id": "todo", "label": "To Do", "order": 0, "group": "planning", "color": "230 99% 67%", "hex": "#536dfe", "hasDirectory": true, "isEntryPoint": true },
+    { "id": "in-progress", "label": "In Progress", "order": 1, "group": "active", "color": "40 100% 50%", "hex": "#ffaa00", "hasDirectory": true },
+    { "id": "done", "label": "Done", "order": 2, "group": "active", "color": "153 100% 39%", "hex": "#00c76d", "hasDirectory": true }
+  ],
+  "edges": [
+    { "from": "todo", "to": "in-progress", "direction": "forward", "command": null, "confirmLevel": "quick", "label": "Start", "description": "Move to in progress.", "dispatchOnly": false },
+    { "from": "in-progress", "to": "done", "direction": "forward", "command": null, "confirmLevel": "quick", "label": "Complete", "description": "Mark as done.", "dispatchOnly": false }
+  ],
+  "hooks": [],
+  "groups": [
+    { "id": "planning", "label": "Planning", "order": 0 },
+    { "id": "active", "label": "Active", "order": 1 }
+  ],
+  "eventInference": [],
+  "allowedCommandPrefixes": ["/scope-", "/git-", "/test-", "/session-"],
+  "terminalStatuses": ["done"],
+  "commitBranchPatterns": "^(main|feat/|fix/|chore/)"
+}

package/templates/quick/rules.md

@@ -0,0 +1,218 @@
+# Non-Negotiable Rules
+
+---
+tokens: ~2K
+load-when: Every coding session
+last-verified: YYYY-MM-DD
+---
+
+Every rule has a **Verify** command. Run it to check compliance.
+
+---
+
+## Code Quality Rules
+
+### Rule 1: No `any` Types
+
+**Rule**: Never use `any` without explicit justification comment
+**Why**: Type safety prevents runtime errors, helps AI understand code
+**Verify**:
+```bash
+grep -r ": any" src --include="*.ts" --include="*.tsx" | grep -v "// justified:"
+```
+**Expected**: No output (or only justified cases)
+**Fix**: Replace with proper type or add `// justified: [reason]`
+
+```typescript
+// ❌ FORBIDDEN
+const data: any = response;
+
+// ✅ REQUIRED
+const data: UserResponse = response;
+
+// ✅ ALLOWED (with justification)
+// justified: third-party SDK returns untyped data
+const result = response as any;
+```
+
+---
+
+### Rule 2: No console.log in Production Code
+
+**Rule**: Never use `console.log`, `console.error`, `console.warn` in production code
+**Why**: No context, leaks to production, unstructured
+**Verify**:
+```bash
+grep -rE "console\.(log|error|warn|info)" src --include="*.ts" --include="*.tsx" | grep -v __tests__
+```
+**Expected**: No output
+**Fix**: Use a structured logger
+
+---
+
+### Rule 3: File Size Limit (400 lines)
+
+**Rule**: Production files must be < 400 lines, tests < 800 lines
+**Why**: Maintainability, cognitive load
+**Verify**:
+```bash
+find src -name "*.ts" -o -name "*.tsx" | xargs wc -l | awk '$1 > 400' | grep -v __tests__
+```
+**Expected**: No output
+**Fix**: Split into focused modules
+
+---
+
+### Rule 4: Max 4 Function Parameters
+
+**Rule**: Functions with >4 parameters must use options object
+**Why**: Readability, maintainability
+**Verify**: Manual review during PR
+
+```typescript
+// ❌ FORBIDDEN
+async function processItem(id: string, name: string, type: string, amount: number, options: object): Promise<Result>
+
+// ✅ REQUIRED
+async function processItem(options: {
+  id: string;
+  name: string;
+  type: string;
+  amount: number;
+  options?: object;
+}): Promise<Result>
+```
+
+---
+
+### Rule 5: Import Ordering
+
+**Rule**: Imports must be: External → Internal (@/) → Relative (./)
+**Why**: Consistency, easier scanning
+**Verify**: ESLint rule (automatic via linter)
+
+```typescript
+// 1. External packages (node_modules)
+import { useState } from 'react';
+import express from 'express';
+
+// 2. Internal aliases (@/)
+import { config } from '@/config';
+import { UserService } from '@/services/user';
+
+// 3. Relative imports (./)
+import { helpers } from './utils';
+```
+
+---
+
+### Rule 6: Explicit Return Types
+
+**Rule**: All exported functions must have explicit return types
+**Why**: Contract clarity, catch errors early
+**Verify**: TypeScript strict mode catches most cases
+
+```typescript
+// ❌ FORBIDDEN
+export async function getUser(id: string) { }
+
+// ✅ REQUIRED
+export async function getUser(id: string): Promise<User | null> { }
+```
+
+---
+
+## Testing Rules
+
+### Rule 7: Tests Required for New Code
+
+**Rule**: All new service and business logic code must have tests
+**Why**: Catch bugs early, documentation
+**Fix**: Create corresponding test file for new modules
+
+---
+
+### Rule 8: Test Structure (Arrange-Act-Assert)
+
+**Rule**: Tests must follow AAA pattern
+**Why**: Readability, maintainability
+
+```typescript
+it('should return user by ID', async () => {
+  // Arrange
+  const userId = 'test-user';
+
+  // Act
+  const result = await getUser(userId);
+
+  // Assert
+  expect(result).toBeDefined();
+  expect(result.id).toBe(userId);
+});
+```
+
+---
+
+## Git Rules
+
+### Rule 9: Quality Gates Before Commit
+
+**Rule**: Configured checks must pass before committing
+**Why**: Prevent broken code in repository
+**Verify**: Run configured commands from orbital.config.json (typeCheck, lint, build, test)
+**Expected**: All pass with exit code 0
+
+---
+
+### Rule 10: One Commit Per Phase
+
+**Rule**: Commit after each phase completion
+**Why**: Progress tracking, easy rollback
+**Fix**: Follow phase-by-phase workflow
+
+---
+
+## Workflow Rules
+
+### Rule 11: Verify Before Claiming Success
+
+**Rule**: Run verification commands BEFORE claiming work is complete. Include output as evidence.
+**Why**: "Should work" is not evidence. Past incidents where untested claims led to broken commits.
+**Fix**: Always run quality gates and show the output before saying "done"
+
+```
+// FORBIDDEN
+"I've fixed the bug, it should work now."
+
+// REQUIRED
+"I've fixed the bug. Here's the verification:
+  type-check: PASS
+  lint: PASS
+  build: PASS
+  tests: 47/47 passing"
+```
+
+---
+
+### Rule 12: Verify Review Feedback Before Implementing
+
+**Rule**: Verify code review feedback against the actual codebase before implementing suggestions.
+**Why**: Review suggestions may be based on stale context or break existing patterns.
+**Fix**: Read the relevant code, confirm the suggestion applies, then implement (or push back with reasoning)
+
+---
+
+## Quick Verification Checklist
+
+Run before every commit:
+
+```bash
+# Run configured quality gates (commands from orbital.config.json)
+# Typical setup:
+# 1. Type check:  npx tsc --noEmit
+# 2. Lint:        npx eslint src/
+# 3. Build:       npm run build
+# 4. Tests:       npm test
+```
+
+All must pass. No exceptions.

package/templates/scopes/_template.md

@@ -0,0 +1,255 @@
+---
+id: NNN
+title: "Scope Title"
+status: planning  # planning | backlog | implementing | review | completed | dev | staging | production
+priority: medium   # critical | high | medium | low
+effort_estimate: "TBD"
+category: "TBD"  # Configure categories in orbital.config.json
+created: YYYY-MM-DD
+updated: YYYY-MM-DD
+spec_locked: false  # true after status = backlog
+blocked_by: []      # scope IDs this depends on
+blocks: []          # scope IDs waiting on this
+tags: []
+sessions: {}  # Auto-populated by skills: {implementScope: [], reviewGate: [], pushToDev: [], ...}
+---
+
+# Scope NNN: Title
+
+═══════════════════════════════════════════════════════════════════
+## PART 1: DASHBOARD
+═══════════════════════════════════════════════════════════════════
+<!--
+PURPOSE: Quick status for user scanning
+UPDATES: Continuously as work progresses
+-->
+
+### Quick Status
+> ⏳ **Status**: Planning | **Phase**: 0 of N | **Spec Locked**: No
+
+### Progress
+| Phase | Description | Status |
+|-------|-------------|--------|
+| 1 | TBD | ⏳ Pending |
+
+### Recent Activity
+- **YYYY-MM-DD HH:MM** - Scope created
+
+### Next Actions
+- [ ] Complete exploration
+- [ ] Draft specification
+- [ ] Get spec approval
+
+═══════════════════════════════════════════════════════════════════
+## PART 2: SPECIFICATION
+═══════════════════════════════════════════════════════════════════
+<!--
+⚠️ FEATURE LOCK: After status = "ready", this section is LOCKED.
+Any agent should be able to implement from ONLY this section.
+Changes after lock require explicit approval + deviation note.
+
+PURPOSE: The authoritative contract for what we're building
+UPDATES: During planning. Frozen after approval.
+-->
+
+### Overview
+
+[Problem statement - what's broken or needed]
+
+**Goal**: [One sentence describing the intended outcome]
+
+### Requirements
+
+#### Must Have
+- [ ] Requirement 1
+- [ ] Requirement 2
+
+#### Nice to Have
+- [ ] Optional enhancement
+
+#### Out of Scope
+- Excluded item 1
+- Excluded item 2
+
+### Technical Approach
+
+[How we're solving it and why this approach was chosen]
+
+**Why this approach**:
+- Reason 1
+- Reason 2
+
+**Architecture** (if applicable):
+```
+[Diagram or structure description]
+```
+
+### Implementation Phases
+
+#### Phase 1: [Name] (estimated time)
+**Objective**: [What this phase accomplishes]
+**Files**: [Files to modify]
+**Changes**: [What changes]
+**Commit**: `type(scope): message`
+**Verification**: [How to verify this phase succeeded]
+
+#### Phase 2: [Name] (estimated time)
+**Objective**: [What this phase accomplishes]
+**Files**: [Files to modify]
+**Changes**: [What changes]
+**Commit**: `type(scope): message`
+**Verification**: [How to verify this phase succeeded]
+
+### Files Summary
+
+| File | Change | Phase |
+|------|--------|-------|
+| `path/to/file.ts` | Description of change | 1 |
+
+### Success Criteria
+
+- [ ] Verifiable condition 1
+- [ ] Verifiable condition 2
+- [ ] Type-check passes
+- [ ] Tests pass
+
+### Risk Assessment
+
+| Risk | Likelihood | Impact | Mitigation |
+|------|------------|--------|------------|
+| Risk description | Low/Med/High | Low/Med/High | How to prevent/handle |
+
+### Definition of Done
+
+- [ ] All phases completed
+- [ ] All success criteria met
+- [ ] All tests passing
+- [ ] Code reviewed
+- [ ] Documentation updated (if applicable)
+
+═══════════════════════════════════════════════════════════════════
+## PART 3: PROCESS
+═══════════════════════════════════════════════════════════════════
+<!--
+PURPOSE: Claude's working memory - exploration, decisions, implementation
+UPDATES: Continuously during work
+DISPLAY: Collapsed by default (user can expand if curious)
+-->
+
+<details>
+<summary>📝 Exploration Log</summary>
+
+<!--
+Record your discovery process here. Each session should include:
+- Trigger: What prompted this exploration
+- Searches: What you looked for and how
+- Findings: What you discovered
+- Insights: What the findings mean
+-->
+
+### Session: YYYY-MM-DD HH:MM
+
+**Trigger**: [What prompted this exploration]
+
+**Search**: [Command or action taken]
+```bash
+# Example search command
+```
+
+**Findings**: [What was found]
+
+**Insight**: [What this means for the solution]
+
+</details>
+
+<details>
+<summary>🤔 Decisions & Reasoning</summary>
+
+<!--
+Capture decisions with alternatives and rationale.
+This helps future sessions understand WHY choices were made.
+-->
+
+### Decisions Made
+
+| # | Decision | Chosen | Rejected Alternatives | Confidence |
+|---|----------|--------|----------------------|------------|
+| 1 | [Decision description] | [What was chosen] | [Alt 1 (why rejected), Alt 2 (why rejected)] | NN% |
+
+### Uncertainties
+
+| Area | Confidence | Mitigation |
+|------|------------|------------|
+| [Area of uncertainty] | NN% | [How to address if wrong] |
+
+### Resolved Questions
+
+| Question | Resolution | Date |
+|----------|------------|------|
+| [Question asked] | [Answer/decision] | YYYY-MM-DD |
+
+</details>
+
+<details>
+<summary>📜 Implementation Log</summary>
+
+<!--
+Updated during implementation. For each phase, record:
+- What was actually done
+- Any issues encountered
+- Actual commit hash
+- Time taken
+- Deviations from spec (if any)
+-->
+
+### Phase 1: [Pending]
+<!-- Example when complete:
+### Phase 1: Completed YYYY-MM-DD HH:MM
+- Added X to file Y
+- Encountered issue with Z, resolved by...
+- Commit: `abc1234`
+- Time: 25 minutes (estimated 30 min)
+-->
+
+</details>
+
+<details>
+<summary>⚠️ Deviations from Spec</summary>
+
+<!--
+If implementation differs from SPECIFICATION, document here:
+- What was specified
+- What was actually done
+- Why the deviation was necessary
+-->
+
+None.
+
+</details>
+
+═══════════════════════════════════════════════════════════════════
+## AGENT REVIEW
+═══════════════════════════════════════════════════════════════════
+<!--
+Populated by /scope-pre-review before implementation begins.
+Contains synthesis of all agent findings.
+-->
+
+### Review Status
+- **Requested**: [agent-list]
+- **Completed**: [agent-list]
+- **Date**: YYYY-MM-DD
+
+### Synthesis
+
+**BLOCKERS** (must fix before implementation):
+- None
+
+**WARNINGS** (should fix):
+- None
+
+**SUGGESTIONS** (nice to have):
+- None
+
+**VERIFIED OK**:
+- None yet

package/templates/settings-hooks.json

@@ -0,0 +1,98 @@
+{
+  "hooks": {
+    "SessionStart": [
+      {
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/init-session.sh", "_orbital": true }
+        ]
+      }
+    ],
+    "SessionEnd": [
+      {
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/end-session.sh", "_orbital": true }
+        ]
+      }
+    ],
+    "PreToolUse": [
+      {
+        "matcher": "Skill",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/git-commit-guard.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/agent-team-gate.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Bash",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/block-push.sh", "_orbital": true },
+{ "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-file-sync.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-lifecycle-gate.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Edit|Write",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-create-gate.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/agent-trigger.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/session-enforcer.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Edit",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/phase-verify-reminder.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/blocker-check.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/dependency-check.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/completion-checklist.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/review-gate-check.sh", "_orbital": true },
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/files-changed-summary.sh", "_orbital": true }
+        ]
+      }
+    ],
+    "PostToolUse": [
+      {
+        "matcher": "Bash",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-commit-logger.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Grep|Glob",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/exploration-logger.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "AskUserQuestion",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/decision-capture.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Edit",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/time-tracker.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Write",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-create-cleanup.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "Skill",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-create-tracker.sh", "_orbital": true }
+        ]
+      },
+      {
+        "matcher": "ExitPlanMode",
+        "hooks": [
+          { "type": "command", "command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/scope-gate.sh", "_orbital": true }
+        ]
+      }
+    ]
+  }
+}

package/templates/skills/git-commit/SKILL.md

@@ -0,0 +1,85 @@
+---
+name: git-commit
+description: Entry point for committing work that routes to the proper git workflow. Use when user says commit, save, or push to ensure correct branch handling.
+user-invocable: true
+---
+
+# /git-commit - Commit Work to Feature Branch
+
+**Use when the user asks to "commit", "save", or similar.**
+
+## Workflow
+
+### Step 0a: Detect Branching Mode
+
+```bash
+BRANCHING_MODE=$(grep '^WORKFLOW_BRANCHING_MODE=' .claude/config/workflow-manifest.sh 2>/dev/null | cut -d'"' -f2)
+[ -z "$BRANCHING_MODE" ] && BRANCHING_MODE="trunk"
+```
+
+### Step 0b: Record Session ID
+
+1. Run: `bash .claude/hooks/get-session-id.sh`
+2. For each scope in `scopes/review/` with a passing verdict:
+   - Append session UUID to `sessions.commit` in frontmatter
+
+### Step 1: Check Branch
+
+```bash
+git branch --show-current
+```
+
+- **Trunk mode**: Allow commits on any branch including `main`
+- **Worktree mode**: Allow commits on the worktree's feature branch
+- **Gitflow mode** (if `BRANCHING_MODE=worktree` and dev/staging/production lists exist): Block if on `main`, `staging`, or `dev` — must be on a feature branch
+
+### Step 2: Scope Transition (local only)
+
+Find scopes in `scopes/review/` that have a passing verdict:
+
+1. List files in `scopes/review/*.md`
+2. For each, extract the scope number and check `.claude/review-verdicts/{NNN}.json`
+3. If verdict exists and `verdict === "PASS"`:
+   - `mv scopes/review/{file} scopes/completed/`
+   - Update frontmatter: `status: completed`
+   - Update DASHBOARD: `📦 **Status**: Committed`
+4. If scope is in `scopes/review/` with **no** passing verdict:
+   - Warn: "Scope {NNN} is in review but hasn't passed the review gate."
+   - Suggest: "Run `/scope-verify {NNN}` before committing."
+   - **Advisory only** — don't block the commit (intermediate commits are fine)
+
+### Step 3: Commit
+
+```bash
+git add <specific code files — scopes are gitignored>
+git commit -m "type(scope): description"
+```
+
+- Stage only code files (scopes/ is gitignored, no need to worry about them)
+- Follow conventional commit format
+- Do NOT push or create PRs — those are separate skills
+
+### Step 4: Signal Completion
+
+If working on a dispatched scope, emit the agent completion event:
+
+```bash
+bash .claude/hooks/orbital-emit.sh AGENT_COMPLETED '{"outcome":"success","action":"save"}' --scope "{NNN}"
+```
+
+## Quick Reference
+
+| User Says | Action |
+|-----------|--------|
+| "commit" | Check branch, then commit |
+| "save" | Same as commit |
+| "push" | Commit first, then advise: `/git-main` to push to main |
+| "push to main" | Commit first, then use `/git-main` |
+| "create PR" | Commit first, then advise: `/git-main` (or `/git-staging` if using Gitflow) |
+| "emergency fix" | Use `/git-hotfix` |
+
+## What This Skill Does NOT Do
+
+- **No push** — use `/git-main` to push/PR to main (or `/git-dev` in Gitflow mode)
+- **No PR creation** — use `/git-main`, `/git-dev`, `/git-staging`, or `/git-production` for PR workflows
+- **No scope moves beyond review→completed** — each lifecycle step is its own skill