luna-agents 2.0.1

package/commands/ll-collab.md

@@ -0,0 +1,87 @@
+---
+name: ll-collab
+displayName: Luna Collab
+description: Team collaboration workflows — standup generation, sprint retros, PR reviews, knowledge sharing, onboarding
+version: 1.0.0
+category: team
+agent: luna-task-planner
+parameters:
+  - name: action
+    type: string
+    description: "Action: standup, retro, review-all, onboard, knowledge, handoff"
+    required: true
+    prompt: true
+mcp_servers:
+  - git
+  - memory
+  - fetch
+  - sequential-thinking
+---
+
+# /collab — Team Intelligence
+
+Automate team workflows that usually eat hours of meetings and coordination.
+
+## Actions
+
+### /collab standup
+```
+Auto-generates daily standup from git activity:
+├── What was merged yesterday (per person)
+├── What's in progress (open PRs, active branches)
+├── Blockers detected (failing CI, stale PRs, dependency conflicts)
+├── Deploy status (what's in staging, what's in prod)
+└── Today's priorities (from sprint board)
+```
+
+### /collab retro
+```
+Auto-generates sprint retrospective:
+���── Velocity: points completed vs planned
+├── Quality: bugs shipped, hotfixes needed
+├── Code health: coverage delta, tech debt change
+├── Highlights: biggest features shipped
+├── Incidents: outages, rollbacks, security issues
+└── Patterns: what's improving, what's degrading
+```
+
+### /collab review-all
+```
+Review all open PRs in one shot:
+├── For each open PR:
+│   ├── Code review (security, quality, tests)
+│   ├── Visual diff (screenshot comparison)
+│   ├── Impact assessment (what it changes)
+│   └── Recommendation (approve, request changes, needs discussion)
+└─�� Priority-ranked summary
+```
+
+### /collab onboard
+```
+Generate onboarding guide for a new team member:
+├── Architecture overview with diagrams
+├── Setup instructions (tested and verified)
+├���─ Key files and entry points explained
+├── Common workflows and conventions
+├── First PR suggestion (good-first-issue)
+└── Team contacts and communication channels
+```
+
+### /collab handoff
+```
+Generate handoff document when leaving a project:
+├── Current state of all features
+├── Known issues and workarounds
+├── Architecture decisions and reasoning
+├── Deployment procedures
+├── Monitoring and alerting setup
+└── Contacts and tribal knowledge
+```
+
+## In Pipes
+
+```bash
+/pipe collab standup >> notify team
+/pipe collab retro >> ghost blog "sprint retrospective"
+/pipe collab review-all >> assert "all PRs approved" >> launch
+```

package/commands/ll-compete.md

@@ -0,0 +1,67 @@
+---
+name: ll-compete
+displayName: Luna Compete
+description: Competitive intelligence — analyze any competitor's product, tech stack, pricing, features, and generate your advantage plan
+version: 1.0.0
+category: strategy
+agent: luna-task-planner
+parameters:
+  - name: competitor
+    type: string
+    description: Competitor URL, name, or GitHub repo
+    required: true
+    prompt: true
+mcp_servers:
+  - fetch
+  - playwright
+  - puppeteer
+  - zai-mcp-server
+  - image-compare
+  - git
+  - sequential-thinking
+  - memory
+---
+
+# /compete — Know Your Enemy, Build Your Advantage
+
+Deep-dive analysis of any competitor. Scrape their product, reverse-engineer their stack, analyze their UX, map their features, and generate your differentiation strategy.
+
+## Analysis Layers
+
+```
+/compete https://linear.app
+              │
+    ┌─────────┼─────────┐
+    ▼         ▼         ▼
+ PRODUCT    TECH      BUSINESS
+ ├ Features  ├ Stack    ├ Pricing tiers
+ ├ UX/UI     ├ Perf     ├ Target market
+ ├ Flows     ├ API      ├ Growth signals
+ ├ Mobile    ├ SEO      ├ Team size
+ └ A11y      └ Infra    └ Funding
+    │         │         │
+    └─────────┼─────────┘
+              ▼
+       DIFFERENTIATION PLAN
+       ├── Features they're missing
+       ├── UX improvements over them
+       ├── Pricing advantage opportunities
+       ├── Technical advantages you can build
+       └── Go-to-market positioning
+```
+
+## Usage
+
+```bash
+/compete https://linear.app                       # Full analysis
+/compete https://n8n.io                            # Workflow competitor
+/compete https://github.com/langchain-ai/langchain # Open source competitor
+/compete "Zapier, Make, n8n"                       # Multi-competitor comparison
+```
+
+## In Pipes
+
+```bash
+/pipe compete "competitor.com" >> idea "build a better version" >> plan >> go *10
+/pipe compete "top 3 competitors" >> brand >> marketing >> launch
+```

package/commands/ll-config-rules.md

@@ -0,0 +1,255 @@
+---
+name: ll-config-rules
+displayName: Luna Persistent Rules
+description: Define persistent project rules that auto-apply to every Luna session, pipe, and agent
+version: 1.0.0
+category: configuration
+agent: luna-task-executor
+parameters:
+  - name: action
+    type: string
+    description: Action (set, get, list, remove, init, export)
+    required: true
+    prompt: true
+workflow:
+  - load_rules_config
+  - execute_action
+  - save_rules_config
+output:
+  - .luna/rules.yaml
+prerequisites: []
+---
+
+# Luna Persistent Rules
+
+Define project-wide rules that **auto-apply to every session, pipe, and agent**.
+
+Stored in `.luna/rules.yaml` — committed to git, shared with your team.
+
+---
+
+## Quick Start
+
+```
+/config-rules init                    # Create .luna/rules.yaml with defaults
+/config-rules set max-lines 100       # Set max file size
+/config-rules set coverage 90         # Set coverage threshold
+/config-rules list                    # Show all active rules
+```
+
+---
+
+## Rules File: `.luna/rules.yaml`
+
+```yaml
+# .luna/rules.yaml — Persistent Luna rules
+# These apply to EVERY Luna session, pipe, and agent automatically
+
+version: 1
+
+# ── Code Quality ──────────────────────────────────────────
+code:
+  max_lines_per_file: 100           # Max lines in any source file
+  max_function_lines: 30            # Max lines per function
+  max_params: 4                     # Max function parameters
+  no_any_types: true                # Disallow TypeScript `any`
+  strict_mode: true                 # TypeScript strict mode
+  naming: camelCase                 # Naming convention (camelCase, snake_case, PascalCase)
+
+# ── Testing ───────────────────────────────────────────────
+testing:
+  unit_coverage: 90                 # Minimum unit test coverage %
+  branch_coverage: 85               # Minimum branch coverage %
+  critical_path_coverage: 100       # Coverage for auth, payments, security
+  require_playwright_e2e: true      # Require Playwright test for every feature
+  require_failing_test_first: true  # Bug fixes must start with a failing test
+  test_naming: describe_it          # Test naming style (describe_it, test_should)
+
+# ── Security ──────────────────────────────────────────────
+security:
+  block_on_critical: true           # Block pipeline on critical vulnerabilities
+  block_on_high: true               # Block pipeline on high vulnerabilities
+  require_input_validation: true    # Require Zod/joi validation on all inputs
+  require_auth_on_routes: true      # All API routes must check auth
+  no_secrets_in_code: true          # Scan for hardcoded secrets
+  require_audit_logging: true       # Auth events, admin actions must be logged
+
+# ── Accessibility ─────────────────────────────────────────
+accessibility:
+  wcag_level: AA                    # WCAG compliance level (A, AA, AAA)
+  require_aria_labels: true         # All interactive elements need ARIA labels
+  min_contrast_ratio: 4.5           # Minimum color contrast ratio
+  require_keyboard_nav: true        # All features must be keyboard accessible
+
+# ── Design ────────────────────────────────────────────────
+design:
+  style_guide: apple-hig            # Design system (apple-hig, material, custom)
+  require_dark_mode: true           # All UI must support dark mode
+  require_responsive: true          # All pages must be responsive
+  max_z_index: 50                   # Prevent z-index wars
+
+# ── Git & Workflow ────────────────────────────────────────
+workflow:
+  commit_style: conventional        # Commit style (conventional, gitmoji, plain)
+  require_pr_review: true           # PRs require code review
+  require_tests_pass: true          # Tests must pass before commit
+  auto_changelog: true              # Auto-generate changelog on release
+  branch_prefix: feature/           # Branch naming prefix
+
+# ── Pipe Defaults ─────────────────────────────────────────
+pipe:
+  default_before: rules             # Always @before:rules on every pipe
+  default_after: test               # Always @after:test on every pipe
+  fail_fast: true                   # Stop pipeline on first failure
+  max_loop_iterations: 10           # Safety cap for *? loops
+  require_approval_for_ship: true   # Always approve before /ship
+
+# ── AI ────────────────────────────────────────────────────
+ai:
+  default_model: opus               # Default AI model for agents
+  max_autonomous_iterations: 5      # Max iterations for /lam and /feature
+  require_review_on_ai_output: true # Auto-review AI-generated code
+
+# ── File Patterns ─────────────────────────────────────────
+patterns:
+  source_files: "src/**/*.{ts,tsx,js,jsx}"
+  test_files: "src/**/*.test.{ts,tsx}"
+  e2e_files: "e2e/**/*.e2e.test.ts"
+  ignore: ["node_modules", "dist", ".next", "coverage"]
+```
+
+---
+
+## Commands
+
+### Initialize rules
+```
+/config-rules init
+```
+Creates `.luna/rules.yaml` with sensible defaults based on your codebase.
+
+### Set a rule
+```
+/config-rules set max-lines 100
+/config-rules set coverage 95
+/config-rules set wcag-level AAA
+/config-rules set commit-style conventional
+/config-rules set require-playwright true
+/config-rules set block-on-critical true
+/config-rules set default-before rules
+/config-rules set default-after test
+/config-rules set require-approval-for-ship true
+```
+
+### Get a rule
+```
+/config-rules get max-lines              # 100
+/config-rules get coverage               # 90
+/config-rules get wcag-level             # AA
+```
+
+### List all rules
+```
+/config-rules list
+```
+Shows all active rules with current values.
+
+### Remove a rule
+```
+/config-rules remove max-function-lines
+```
+
+### Export rules
+```
+/config-rules export                     # Print rules.yaml to console
+/config-rules export clipboard           # Copy to clipboard
+```
+
+---
+
+## How Rules Auto-Apply
+
+### Every Luna Session
+When you start any Luna command, rules are loaded from `.luna/rules.yaml`:
+- `/go` respects `max_lines_per_file`, `no_any_types`, `strict_mode`
+- `/rev` checks against all code quality and security rules
+- `/test` enforces coverage thresholds
+- `/ship` requires approval if `require_approval_for_ship: true`
+
+### Every Pipe
+Pipe defaults from `rules.yaml` are injected automatically:
+```yaml
+pipe:
+  default_before: rules
+  default_after: test
+```
+This means:
+```
+/pipe go *5 >> rev >> ship
+```
+Automatically becomes:
+```
+/pipe @before:rules @after:test go *5 >> rev >> approve "Ship?" >> ship
+```
+
+### Every Assert
+Assertions read thresholds from rules:
+```
+/pipe assert coverage    # Uses testing.unit_coverage from rules.yaml
+/pipe assert file-size   # Uses code.max_lines_per_file from rules.yaml
+/pipe assert security    # Uses security.block_on_critical from rules.yaml
+```
+
+### AI Agents
+AI agents respect:
+```yaml
+ai:
+  default_model: opus
+  max_autonomous_iterations: 5
+  require_review_on_ai_output: true
+```
+
+---
+
+## Rule Inheritance
+
+Rules cascade:
+```
+Portfolio CLAUDE.md (200-line cap, 90% coverage)
+  └── Project .luna/rules.yaml (100-line cap, 95% coverage)  ← stricter, overrides
+       └── Session /rules (runtime overrides)                 ← strictest
+```
+
+**Rules can only get stricter, never weaker.**
+
+- Portfolio says 200 lines → project can set 100, not 300
+- Portfolio says 90% coverage → project can set 95%, not 80%
+
+---
+
+## Team Sharing
+
+`.luna/rules.yaml` is a regular file — commit it to git:
+
+```bash
+git add .luna/rules.yaml
+git commit -m "chore: set project rules — 100-line cap, 95% coverage"
+```
+
+Everyone on the team gets the same rules automatically.
+
+---
+
+## Presets
+
+```
+/config-rules init strict              # Strictest settings
+/config-rules init standard            # Balanced defaults
+/config-rules init startup             # Fast-moving, fewer gates
+```
+
+| Preset | Max Lines | Coverage | Security | Approval |
+|--------|-----------|----------|----------|----------|
+| `strict` | 80 | 95% | Block all | Required |
+| `standard` | 100 | 90% | Block critical | Required |
+| `startup` | 150 | 80% | Warn only | Optional |