claude-skill-lord 2.0.4 → 2.0.6

package/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "claude-skill-lord",
-  "version": "2.0.4",
-  "description": "Curated Claude Code plugin — 44 agents, 170 skills, 115 commands, 13 language rules with intelligent skill routing",
+  "version": "2.0.6",
+  "description": "Curated Claude Code plugin — 43 agents, 170 skills, 114 commands, 11 language rules",
   "author": {
     "name": "Dong Anh",
     "email": "donganhvu20@gmail.com"
@@ -14,7 +14,6 @@
     "agents",
     "skills",
     "commands",
-    "skill-routing",
     "quality-gate",
     "tdd",
     "code-review",
@@ -62,7 +61,6 @@
     "./agents/scout-external.md",
     "./agents/scout.md",
     "./agents/security-reviewer.md",
-    "./agents/skill-router.md",
     "./agents/tdd-guide.md",
     "./agents/tester.md",
     "./agents/typescript-reviewer.md",

package/CLAUDE.md

@@ -39,7 +39,6 @@ Curated best-of-both plugin merging ClaudeKit Engineer (base) + Everything Claud
 | loop-operator | Autonomous workflows |
 | chief-of-staff | Multi-channel coordination |
 | harness-optimizer | Self-optimization |
-| skill-router | Advisory skill recommendation |
 | quality-gate | Output validation |
 | researcher | Deep research agent |
 | tester | Test execution agent |

package/README.md

@@ -4,16 +4,16 @@
 
 [![npm](https://img.shields.io/npm/v/claude-skill-lord)](https://www.npmjs.com/package/claude-skill-lord)
 ![License](https://img.shields.io/badge/license-MIT-green)
-![Agents](https://img.shields.io/badge/agents-44-purple)
+![Agents](https://img.shields.io/badge/agents-43-purple)
 ![Skills](https://img.shields.io/badge/skills-170-orange)
-![Commands](https://img.shields.io/badge/commands-115-red)
+![Commands](https://img.shields.io/badge/commands-114-red)
 ![Rules](https://img.shields.io/badge/rules-11%20languages-blue)
 
 ---
 
 ## Why Claude Skill Lord?
 
-- **Stop configuring, start building** — 170 skills, 44 agents, 115 commands work out of the box
+- **Stop configuring, start building** — 170 skills, 43 agents, 114 commands work out of the box
 - **Multi-language support** — 11 language-specific rule sets (TypeScript, Python, Go, Rust, Java, Kotlin, C++, C#, PHP, Perl, Swift) + common rules
 - **Language-specific agents** — dedicated reviewers and build resolvers for 8 languages
 - **Design intelligence built-in** — 67 UI styles, 161 color palettes, reasoning engine for production-grade design decisions
@@ -78,14 +78,13 @@ node scripts/sl.js init full --target /path/to/your/project
 ```mermaid
 graph TD
     subgraph Plugin["Claude Skill Lord Plugin"]
-        A["Agents (44)"]
-        B["Commands (115)"]
+        A["Agents (43)"]
+        B["Commands (114)"]
         C["Skills (170)"]
-        R["Rules (13 langs)"]
+        R["Rules (11 langs)"]
     end
 
     subgraph Intelligence["Intelligence Layer"]
-        D["Skill Router"]
         E["Quality Gate"]
     end
 
@@ -96,15 +95,14 @@ graph TD
         CTX["Contexts"]
     end
 
-    A --> D
-    B --> D
-    C --> D
-    R --> D
-    D --> E
+    A --> E
+    B --> E
+    C --> E
+    R --> E
     E --> F
     E --> G
     E --> H
-    CTX --> D
+    CTX --> E
 ```
 
 ---
@@ -193,12 +191,11 @@ The most common workflow — plan first, then implement, then validate:
 | `/cook` | Implement features end-to-end |
 | `/tdd` | Test-driven development workflow |
 | `/debug` | Deep root-cause investigation |
-| `/design` | Create UI designs (variants: `fast`, `good`, `3d`, `screenshot`, `video`, `describe`) |
-| `/route` | Get skill recommendations for your task |
+| `/design:good` | Create UI designs (variants: `fast`, `good`, `3d`, `screenshot`, `video`, `describe`) |
 | `/audit` | Run quality gate checks |
 
 <details>
-<summary><strong>All commands (115)</strong></summary>
+<summary><strong>All commands (114)</strong></summary>
 
 | Category | Commands |
 |----------|----------|
@@ -217,7 +214,7 @@ The most common workflow — plan first, then implement, then validate:
 | **Skills** | `/skill:add`, `/skill:create`, `/skill:optimize`, `/skill:fix-logs`, `/skill-create`, `/skill-health` |
 | **Quality** | `/quality-gate`, `/audit`, `/refactor-clean`, `/prompt-optimize` |
 | **Loop** | `/loop-start`, `/loop-status` |
-| **Other** | `/brainstorm`, `/learn`, `/evolve`, `/model-route`, `/route`, `/ask`, `/journal`, `/watzup` |
+| **Other** | `/brainstorm`, `/learn`, `/evolve`, `/model-route`, `/ask`, `/journal`, `/watzup` |
 
 See [commands/](commands/) for the full list.
 
@@ -228,7 +225,7 @@ See [commands/](commands/) for the full list.
 ## Agents
 
 <details>
-<summary><strong>44 agents — click to expand</strong></summary>
+<summary><strong>43 agents — click to expand</strong></summary>
 
 ### Core Agents
 
@@ -258,7 +255,6 @@ See [commands/](commands/) for the full list.
 | loop-operator | Autonomous development loops |
 | chief-of-staff | Multi-channel coordination |
 | harness-optimizer | Agent self-optimization |
-| skill-router | Advisory skill recommendations |
 | quality-gate | Output validation |
 | researcher | Deep research agent |
 | tester | Test execution agent |
@@ -304,7 +300,7 @@ All skills live in `./skills/<name>/SKILL.md` — flat structure.
 debugging, code-review, tdd-workflow, testing, backend-development, frontend-development, web-frameworks, ui-styling, databases, api-design, devops, sequential-thinking, research, planning, problem-solving, coding-standards
 
 ### Frontend & Design
-ui-ux-pro-max, react-best-practices, frontend-patterns, frontend-design, frontend-slides, design, design-system, brand, banner-design, slides, aesthetic, web-design-guidelines, liquid-glass-design, threejs
+ui-ux-pro-max, react-best-practices, frontend-patterns, frontend-design, frontend-slides, nextjs-turbopack, design, design-system, brand, banner-design, slides, aesthetic, web-design-guidelines, liquid-glass-design, threejs
 
 ### Backend & API
 backend-patterns, api-design, mcp-server-patterns, mcp-management, mcp-builder
@@ -313,7 +309,7 @@ backend-patterns, api-design, mcp-server-patterns, mcp-management, mcp-builder
 python-patterns, golang-patterns, rust-patterns, kotlin-patterns, perl-patterns, django-patterns, laravel-patterns, springboot-patterns, swiftui-patterns, nuxt4-patterns
 
 ### Language Testing & Security
-python-testing, golang-testing, rust-testing, kotlin-testing, cpp-testing, django-tdd, laravel-tdd, springboot-tdd, django-security, laravel-security, springboot-security, django-verification, laravel-verification, springboot-verification
+python-testing, golang-testing, rust-testing, kotlin-testing, cpp-testing, perl-testing, perl-security, django-tdd, laravel-tdd, springboot-tdd, django-security, laravel-security, springboot-security, django-verification, laravel-verification, springboot-verification
 
 ### Language Specialized
 kotlin-coroutines-flows, kotlin-exposed-patterns, kotlin-ktor-patterns, java-coding-standards, cpp-coding-standards, swift-actor-persistence, swift-concurrency-6-2, swift-protocol-di-testing, jpa-patterns, compose-multiplatform-patterns
@@ -331,7 +327,7 @@ postgres-patterns, database-migrations, clickhouse-io
 ai-multimodal, pytorch-patterns, google-adk-python, cost-aware-llm-pipeline, foundation-models-on-device, prompt-optimizer
 
 ### Agentic Engineering
-agentic-engineering, agent-harness-construction, agent-eval, autonomous-loops, continuous-agent-loop, continuous-learning, continuous-learning-v2, eval-harness, verification-loop, enterprise-agent-ops
+agentic-engineering, ai-first-engineering, ai-regression-testing, agent-harness-construction, agent-eval, autonomous-loops, continuous-agent-loop, continuous-learning, continuous-learning-v2, eval-harness, verification-loop, enterprise-agent-ops
 
 ### Content & Business
 article-writing, content-engine, crosspost, market-research, investor-outreach, investor-materials, shopify
@@ -343,10 +339,10 @@ security-review, security-scan, better-auth, payment-integration, safety-guard
 media-processing, video-editing, videodb, fal-ai-media, nutrient-document-processing
 
 ### Tools & Utilities
-repomix, chrome-devtools, docs-seeker, documentation-lookup, skill-creator, data-scraper-agent, exa-search, x-api, bun-runtime, nanoclaw-repl, dmux-workflows
+repomix, chrome-devtools, docs-seeker, documentation-lookup, document-skills, skill-creator, data-scraper-agent, exa-search, x-api, bun-runtime, nanoclaw-repl, dmux-workflows
 
 ### Research & Strategy
-deep-research, strategic-compact, search-first, iterative-retrieval, codebase-onboarding, blueprint, santa-method, team-builder
+deep-research, strategic-compact, search-first, iterative-retrieval, codebase-onboarding, blueprint, santa-method, team-builder, architecture-decision-records
 
 ### Domain-Specific
 carrier-relationship-management, customs-trade-compliance, energy-procurement, inventory-demand-planning, logistics-exception-management, production-scheduling, quality-nonconformance, returns-reverse-logistics, visa-doc-translate

package/manifests/install-modules.json

@@ -4,7 +4,7 @@
     {
       "id": "agents",
       "kind": "agents",
-      "description": "All 44 agents — planner, architect, code-reviewer, security, language-specific reviewers & build resolvers, and more",
+      "description": "All 43 agents — planner, architect, code-reviewer, security, language-specific reviewers & build resolvers, and more",
       "paths": ["agents/"],
       "defaultInstall": true,
       "cost": "medium",
@@ -13,7 +13,7 @@
     {
       "id": "commands",
       "kind": "commands",
-      "description": "All 115 commands — plan, code, test, fix, cook, design, bootstrap, multi-*, session, build, review, and more",
+      "description": "All 114 commands — plan, code, test, fix, cook, design, bootstrap, multi-*, session, build, review, and more",
       "paths": ["commands/"],
       "defaultInstall": true,
       "cost": "light",

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "claude-skill-lord",
-  "version": "2.0.4",
-  "description": "Curated Claude Code plugin — 44 agents, 170 skills, 115 commands, 13 language rules with intelligent skill routing",
+  "version": "2.0.6",
+  "description": "Curated Claude Code plugin — 43 agents, 170 skills, 114 commands, 11 language rules",
   "author": {
     "name": "Dong Anh",
     "email": "donganhvu20@gmail.com"
@@ -22,7 +22,6 @@
     "commands",
     "ai",
     "developer-tools",
-    "skill-routing",
     "quality-gate",
     "tdd",
     "code-review"

package/scripts/install.js

@@ -115,6 +115,19 @@ const pluginJson = buildPluginJson('full', filesToCopy, pkg.version);
 fs.writeFileSync(pluginJsonPath, JSON.stringify(pluginJson, null, 2));
 console.log(`\n  Generated: .claude/plugin.json`);
 
+// Copy CLAUDE.md to project root (Claude Code reads this automatically)
+const claudeMdSrc = path.join(skillLordRoot, 'CLAUDE.md');
+const claudeMdDest = path.join(targetDir, 'CLAUDE.md');
+if (fs.existsSync(claudeMdSrc)) {
+  if (fs.existsSync(claudeMdDest)) {
+    console.log(`  Skipped:  CLAUDE.md (already exists in project root)`);
+  } else {
+    fs.copyFileSync(claudeMdSrc, claudeMdDest);
+    console.log(`  Copied:   CLAUDE.md → project root`);
+    copied++;
+  }
+}
+
 console.log(`\n  Installation complete!`);
 console.log(`  Copied:  ${copied} files`);
 if (skipped > 0) console.log(`  Skipped: ${skipped} files (already exist)`);

package/scripts/sl.js

@@ -184,6 +184,28 @@ const commands = {
       console.log(`  Rebuilt plugin.json (${rebuilt.agents.length} agents, ${rebuilt.skills.length} skill dirs)`);
     }
 
+    // Migrate CLAUDE.md to project root
+    const claudeMdSrc = path.join(rootDir, 'CLAUDE.md');
+    const claudeMdDest = path.join(targetDir, 'CLAUDE.md');
+    if (fs.existsSync(claudeMdSrc)) {
+      if (!fs.existsSync(claudeMdDest)) {
+        if (dryRun) console.log(`  + NEW: CLAUDE.md (project root)`);
+        else {
+          fs.copyFileSync(claudeMdSrc, claudeMdDest);
+          console.log(`  Copied CLAUDE.md → project root`);
+        }
+        added++;
+      } else {
+        const srcBuf = fs.readFileSync(claudeMdSrc);
+        const destBuf = fs.readFileSync(claudeMdDest);
+        if (!srcBuf.equals(destBuf)) {
+          if (dryRun) console.log(`  ~ UPDATED: CLAUDE.md (project root)`);
+          else fs.copyFileSync(claudeMdSrc, claudeMdDest);
+          updated++;
+        }
+      }
+    }
+
     console.log(`\n  ${dryRun ? '[DRY RUN] ' : ''}Results:`);
     console.log(`    New files:     ${added}`);
     console.log(`    Updated:       ${updated}`);
@@ -269,9 +291,6 @@ const commands = {
   },
 
   list: () => {
-    const manifest = JSON.parse(
-      fs.readFileSync(path.join(rootDir, 'skills', 'manifest.json'), 'utf8')
-    );
     const agents = fs.readdirSync(path.join(rootDir, 'agents')).filter(f => f.endsWith('.md'));
 
     console.log(`\n  Claude Skill Lord Components\n`);
@@ -279,12 +298,12 @@ const commands = {
     console.log(`  Agents: ${agents.length}`);
     agents.forEach(a => console.log(`    - ${a.replace('.md', '')}`));
 
-    console.log(`\n  Skills: ${manifest.skills.length}`);
-    [1, 2, 3].forEach(tier => {
-      const skills = manifest.skills.filter(s => s.tier === tier);
-      console.log(`\n  Tier ${tier} (${skills.length}):`);
-      skills.forEach(s => console.log(`    - ${s.name}: ${s.description}`));
-    });
+    const skillsDir = path.join(rootDir, 'skills');
+    const skills = fs.readdirSync(skillsDir, { withFileTypes: true })
+      .filter(e => e.isDirectory() && fs.existsSync(path.join(skillsDir, e.name, 'SKILL.md')))
+      .map(e => e.name);
+    console.log(`\n  Skills: ${skills.length}`);
+    skills.forEach(s => console.log(`    - ${s}`));
 
     const cmds = [];
     const collectCmds = (dir) => {
@@ -362,9 +381,11 @@ const commands = {
       if (missing.length) throw new Error(`missing: ${missing.join(', ')}`);
     });
 
-    check('manifest.json valid', () => {
-      const m = JSON.parse(fs.readFileSync(path.join(rootDir, 'skills', 'manifest.json'), 'utf8'));
-      if (!m.skills || m.skills.length === 0) throw new Error('no skills');
+    check('Skills directory populated', () => {
+      const skillsDir = path.join(rootDir, 'skills');
+      const skills = fs.readdirSync(skillsDir, { withFileTypes: true })
+        .filter(e => e.isDirectory() && fs.existsSync(path.join(skillsDir, e.name, 'SKILL.md')));
+      if (skills.length === 0) throw new Error('no skills with SKILL.md found');
     });
 
     check('hooks.json valid', () => {
@@ -433,7 +454,7 @@ const commands = {
   Usage: csl <command> [options]
 
   Commands:
-    init                Install in current project (22 agents, 61 skills, all commands)
+    init                Install in current project (43 agents, 170 skills, 114 commands)
     update              Update CLI to latest version
     migrate             Update project files after csl update
     diff                Compare project files with source package

package/skills/ai-multimodal/.env.example

@@ -0,0 +1,97 @@
+# Google Gemini API Configuration
+
+# ============================================================================
+# OPTION 1: Google AI Studio (Default - Recommended for most users)
+# ============================================================================
+# Get your API key: https://aistudio.google.com/apikey
+GEMINI_API_KEY=your_api_key_here
+
+# ============================================================================
+# OPTION 2: Vertex AI (Google Cloud Platform)
+# ============================================================================
+# Uncomment these lines to use Vertex AI instead of Google AI Studio
+# GEMINI_USE_VERTEX=true
+# VERTEX_PROJECT_ID=your-gcp-project-id
+# VERTEX_LOCATION=us-central1
+
+# ============================================================================
+# Model Selection (Optional)
+# ============================================================================
+# Override default model for specific tasks
+# Default: gemini-2.5-flash for most tasks
+# GEMINI_MODEL=gemini-2.5-flash
+# GEMINI_IMAGE_GEN_MODEL=gemini-2.5-flash-image
+
+# ============================================================================
+# Rate Limiting Configuration (Optional)
+# ============================================================================
+# Requests per minute limit (adjust based on your tier)
+# GEMINI_RPM_LIMIT=15
+
+# Tokens per minute limit
+# GEMINI_TPM_LIMIT=4000000
+
+# Requests per day limit
+# GEMINI_RPD_LIMIT=1500
+
+# ============================================================================
+# Processing Options (Optional)
+# ============================================================================
+# Video resolution mode: default or low-res
+# low-res uses ~100 tokens/second vs ~300 for default
+# GEMINI_VIDEO_RESOLUTION=default
+
+# Audio quality: default (16 Kbps mono, auto-downsampled)
+# GEMINI_AUDIO_QUALITY=default
+
+# PDF processing mode: inline (<20MB) or file-api (>20MB, automatic)
+# GEMINI_PDF_MODE=auto
+
+# ============================================================================
+# Retry Configuration (Optional)
+# ============================================================================
+# Maximum retry attempts for failed requests
+# GEMINI_MAX_RETRIES=3
+
+# Initial retry delay in seconds (uses exponential backoff)
+# GEMINI_RETRY_DELAY=1
+
+# ============================================================================
+# Output Configuration (Optional)
+# ============================================================================
+# Default output directory for generated images
+# OUTPUT_DIR=./output
+
+# Image output format (png or jpeg)
+# IMAGE_FORMAT=png
+
+# Image quality for JPEG (1-100)
+# IMAGE_QUALITY=95
+
+# ============================================================================
+# Context Caching (Optional)
+# ============================================================================
+# Enable context caching for repeated queries on same file
+# GEMINI_ENABLE_CACHING=true
+
+# Cache TTL in seconds (default: 1800 = 30 minutes)
+# GEMINI_CACHE_TTL=1800
+
+# ============================================================================
+# Logging (Optional)
+# ============================================================================
+# Log level: DEBUG, INFO, WARNING, ERROR, CRITICAL
+# LOG_LEVEL=INFO
+
+# Log file path
+# LOG_FILE=./logs/gemini.log
+
+# ============================================================================
+# Notes
+# ============================================================================
+# 1. Never commit API keys to version control
+# 2. Add .env to .gitignore
+# 3. API keys can be restricted in Google Cloud Console
+# 4. Monitor usage at: https://aistudio.google.com/apikey
+# 5. Free tier limits: 15 RPM, 1M-4M TPM, 1,500 RPD
+# 6. Vertex AI requires GCP authentication via gcloud CLI

package/skills/devops/.env.example

@@ -0,0 +1,76 @@
+# DevOps Skill - Environment Variables
+
+# =============================================================================
+# Cloudflare Configuration
+# =============================================================================
+# Get these from: https://dash.cloudflare.com
+# API Token: Profile -> API Tokens -> Create Token
+# Account ID: Overview -> Account ID (right sidebar)
+
+CLOUDFLARE_API_TOKEN=your_cloudflare_api_token_here
+CLOUDFLARE_ACCOUNT_ID=your_cloudflare_account_id_here
+
+# Optional: Specific zone configuration
+# CLOUDFLARE_ZONE_ID=your_zone_id_here
+
+# =============================================================================
+# Google Cloud Configuration
+# =============================================================================
+# Authentication via service account key file or gcloud CLI
+# Download from: IAM & Admin -> Service Accounts -> Create Key
+
+# Option 1: Service account key file path
+GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account-key.json
+
+# Option 2: Project configuration
+# GCP_PROJECT_ID=your-project-id
+# GCP_REGION=us-central1
+# GCP_ZONE=us-central1-a
+
+# =============================================================================
+# Docker Configuration
+# =============================================================================
+# Optional: Docker registry authentication
+
+# Docker Hub
+# DOCKER_USERNAME=your_docker_username
+# DOCKER_PASSWORD=your_docker_password
+
+# Google Container Registry (GCR)
+# GCR_HOSTNAME=gcr.io
+# GCR_PROJECT_ID=your-project-id
+
+# AWS ECR
+# AWS_ACCOUNT_ID=123456789012
+# AWS_REGION=us-east-1
+
+# =============================================================================
+# CI/CD Configuration
+# =============================================================================
+# Optional: For automated deployments
+
+# GitHub Actions
+# GITHUB_TOKEN=your_github_token
+
+# GitLab CI
+# GITLAB_TOKEN=your_gitlab_token
+
+# =============================================================================
+# Monitoring & Logging
+# =============================================================================
+# Optional: For observability
+
+# Sentry
+# SENTRY_DSN=your_sentry_dsn
+
+# Datadog
+# DD_API_KEY=your_datadog_api_key
+
+# =============================================================================
+# Notes
+# =============================================================================
+# 1. Copy this file to .env and fill in your actual values
+# 2. Never commit .env file to version control
+# 3. Use different credentials for dev/staging/production
+# 4. Rotate credentials regularly
+# 5. Use least-privilege principle for API tokens

package/skills/docs-seeker/.env.example

@@ -0,0 +1,15 @@
+# Context7 API Configuration (optional)
+# Get your API key from https://context7.com/dashboard/api-keys
+CONTEXT7_API_KEY=
+
+# Gemini API Configuration (optional, for ai-multimodal integration)
+# Get your API key from https://aistudio.google.com/app/apikey
+GEMINI_API_KEY=
+
+# GitHub Token (optional, for higher rate limits on repository analysis)
+# Create at https://github.com/settings/tokens
+GITHUB_TOKEN=
+
+# Output settings
+OUTPUT_FORMAT=json
+DEBUG=false

package/skills/payment-integration/scripts/.env.example

@@ -0,0 +1,20 @@
+# SePay Configuration
+SEPAY_MERCHANT_ID=SP-TEST-XXXXXXX
+SEPAY_SECRET_KEY=spsk_test_xxxxxxxxxxxxx
+SEPAY_ENV=sandbox  # or 'production'
+
+# SePay Webhook Configuration
+SEPAY_WEBHOOK_AUTH_TYPE=api_key  # or 'oauth2' or 'none'
+SEPAY_WEBHOOK_API_KEY=your_webhook_api_key
+
+# Polar Configuration
+POLAR_ACCESS_TOKEN=polar_xxxxxxxxxxxxxxxx
+POLAR_SERVER=sandbox  # or 'production'
+POLAR_ORG_ID=org_xxxxxxxxxxxxx
+
+# Polar Webhook Configuration
+POLAR_WEBHOOK_SECRET=base64_encoded_secret
+
+# Optional: Database or other configuration
+# DATABASE_URL=postgresql://user:password@localhost:5432/dbname
+# REDIS_URL=redis://localhost:6379

package/skills/sequential-thinking/.env.example

@@ -0,0 +1,8 @@
+# Sequential Thinking Configuration
+
+# Disable thought logging output (useful for automated processing)
+# Set to "true" to disable console logging
+DISABLE_THOUGHT_LOGGING=false
+
+# History file location (optional, defaults to scripts/.thought-history.json)
+# THOUGHT_HISTORY_FILE=/path/to/custom/history.json

package/agents/skill-router.md

@@ -1,112 +0,0 @@
----
-name: skill-router
-description: Advisory skill recommendation agent — analyzes user tasks and recommends relevant skills from the SkillLord catalog
-tools: ["Glob", "Grep", "Read"]
-model: sonnet
----
-
-# Skill Router (Advisory)
-
-You are an advisory skill routing agent for SkillLord. Your job is to analyze the user's request and recommend the most relevant skills from the catalog. You do NOT load or execute skills — you only recommend.
-
-## Step-by-Step Process
-
-### Step 1: Load the Skill Catalog
-Read `skills/manifest.json` at the plugin root. This file contains all 55 skills with metadata:
-```json
-{ "name": "...", "tier": 1|2|3, "path": "...", "tags": [...], "description": "...", "dependencies": [...] }
-```
-
-### Step 2: Analyze the User Request
-Extract from the user's request:
-- **Intent**: What action? (build, fix, debug, test, deploy, design, review, plan, research)
-- **Technology stack**: What tech? (React, Python, Go, PostgreSQL, Docker, etc.)
-- **Domain**: What area? (frontend, backend, database, security, DevOps, mobile, AI/ML)
-- **Task type**: New feature, bug fix, refactor, migration, optimization, testing
-
-### Step 3: Score Each Skill
-
-For each skill in the manifest, calculate a relevance score (0.0 to 1.0):
-
-**Tag Match (40% weight):**
-- Extract keywords from user request (lowercase, split by spaces/punctuation)
-- For each skill tag that matches a request keyword: +1.0
-- For each skill tag that partially matches (substring): +0.5
-- Normalize: `tag_score = matches / total_tags`
-
-**Description Match (30% weight):**
-- Count keyword hits in the skill's description
-- Normalize: `desc_score = hits / total_keywords`
-
-**Tier Preference (20% weight):**
-- Tier 1: `tier_score = 1.0`
-- Tier 2: `tier_score = 0.6`
-- Tier 3: `tier_score = 0.3`
-
-**Dependency Bonus (10% weight):**
-- If a recommended skill is a dependency of another recommended skill: +0.5
-- If skills form a logical chain: +0.3
-
-**Final Score:**
-```
-score = (tag_score * 0.4) + (desc_score * 0.3) + (tier_score * 0.2) + (dep_bonus * 0.1)
-```
-
-### Step 4: Rank and Filter
-1. Sort skills by score descending
-2. Filter out skills with score < 0.3
-3. Take top 7 results
-4. Ensure at least 1 Tier-1 skill is included (if any scored > 0.2)
-
-### Step 5: Output Recommendations
-
-Present results in this format:
-
-```
-## Skill Recommendations
-
-Based on your request: "{user_request}"
-
-| # | Skill | Tier | Score | Reason |
-|---|-------|------|-------|--------|
-| 1 | debugging | T1 | 0.95 | Direct match: debug, error, fix tags |
-| 2 | backend-development | T1 | 0.82 | Stack match: API, server, Node.js |
-| 3 | testing | T1 | 0.71 | Related: test coverage for fixes |
-| 4 | security-patterns | T1 | 0.55 | Context: auth-related debugging |
-| 5 | better-auth | T2 | 0.48 | Domain: authentication framework |
-
-### How to Use
-These skills are already available in your SkillLord installation.
-The most relevant skills will be automatically considered by Claude
-when processing your requests. You can also explicitly reference
-a skill's patterns by mentioning it in your prompt.
-
-### Skill Dependencies
-- `e2e-testing` requires familiarity with `testing`
-- `postgres-patterns` builds on `databases`
-```
-
-## Example Routing
-
-**Request:** "I need to add OAuth2 login to my Next.js app with Google provider"
-
-**Analysis:**
-- Intent: build (new feature)
-- Stack: Next.js, OAuth2, Google
-- Domain: frontend + auth
-- Type: new feature
-
-**Top recommendations:**
-1. `better-auth` (T2, 0.95) — OAuth framework patterns
-2. `frontend-development` (T1, 0.85) — Next.js patterns
-3. `web-frameworks` (T1, 0.78) — Next.js App Router
-4. `security-patterns` (T1, 0.65) — Auth security best practices
-5. `api-design` (T1, 0.52) — API route design for auth endpoints
-
-## Constraints
-
-- **ADVISORY ONLY**: You cannot dynamically load or inject skills into the session
-- **Max 7 recommendations**: Keep output focused, not exhaustive
-- **Show reasoning**: Always explain WHY each skill is recommended
-- **Be honest**: If no skills match well, say so. Don't force recommendations
-- **Tier awareness**: Mention if recommended skills are in Tier 2/3 (user may need to install full profile)