Add 260+ Claude Code skills from skills.sh

Complete collection of AI agent skills including: - Frontend Development (Vue, React, Next.js, Three.js) - Backend Development (NestJS, FastAPI, Node.js) - Mobile Development (React Native, Expo) - Testing (E2E, frontend, webapp) - DevOps (GitHub Actions, CI/CD) - Marketing (SEO, copywriting, analytics) - Security (binary analysis, vulnerability scanning) - And many more... Synchronized from: https://skills.sh/ Co-Authored-By: Claude <noreply@anthropic.com>
2026-01-23 18:02:28 +00:00
commit 07242683bf
3300 changed files with 1223105 additions and 0 deletions
--- a/agent-pipeline-builder/.triggers/keywords.json
+++ b/agent-pipeline-builder/.triggers/keywords.json
@@ -0,0 +1,28 @@
+{
+  "skills": [
+    {
+      "name": "agent-pipeline-builder",
+      "triggers": [
+        "multi-agent pipeline",
+        "agent pipeline",
+        "multi agent workflow",
+        "create pipeline",
+        "build pipeline",
+        "orchestrate agents",
+        "agent workflow",
+        "pipeline architecture",
+        "sequential agents",
+        "agent chain",
+        "data pipeline",
+        "agent orchestration",
+        "multi-stage workflow",
+        "agent composition",
+        "pipeline pattern",
+        "researcher analyzer writer",
+        "funnel pattern",
+        "transformation pipeline",
+        "agent data flow"
+      ]
+    }
+  ]
+}
--- a/agent-pipeline-builder/SKILL.md
+++ b/agent-pipeline-builder/SKILL.md
@@ -0,0 +1,357 @@
+---
+name: agent-pipeline-builder
+description: Build multi-agent pipelines with structured data flow between agents. Use when creating workflows where each agent has a specialized role and passes output to the next agent.
+allowed-tools: Write, Edit, Read, Bash, WebSearch
+license: MIT
+---
+
+# Agent Pipeline Builder
+
+Build reliable multi-agent workflows where each agent has a single, focused responsibility and outputs structured data that the next agent consumes.
+
+## When to Use This Skill
+
+Use this skill when:
+- Building complex workflows that need multiple specialized agents
+- Creating content pipelines (research → analysis → writing)
+- Designing data processing flows with validation at each stage
+- Implementing "funnel" patterns where broad input becomes focused output
+
+## Pipeline Pattern
+
+A pipeline consists of:
+1. **Stage 1: Researcher/Gatherer** - Fetches raw data (WebSearch, file reading, API calls)
+2. **Stage 2: Analyzer/Filter** - Processes and selects best options
+3. **Stage 3: Creator/Writer** - Produces final output
+
+Each stage:
+- Has ONE job
+- Outputs structured JSON (or YAML)
+- Wraps output in markers (e.g., `<<<stage>>>...<<<end-stage>>>`)
+- Passes data to next stage via stdin or file
+
+## RalphLoop "Tackle Until Solved" Integration
+
+For complex pipelines (3+ stages or complexity >= 5), agent-pipeline-builder automatically delegates to Ralph Orchestrator for autonomous pipeline construction and testing.
+
+### When Ralph is Triggered
+
+Ralph mode activates for pipelines with:
+- 3 or more stages
+- Complex stage patterns (external APIs, complex processing, conditional logic)
+- Parallel stage execution
+- User opt-in via `RALPH_AUTO=true` or `PIPELINE_USE_RALPH=true`
+
+### Using Ralph Integration
+
+When a complex pipeline is detected:
+
+1. Check for Python integration module:
+   ```bash
+   python3 /home/uroma/.claude/skills/agent-pipeline-builder/ralph-pipeline.py --test-complexity
+   ```
+
+2. If complex, delegate to Ralph:
+   ```bash
+   /home/uroma/obsidian-web-interface/bin/ralphloop -i .ralph/PIPELINE.md
+   ```
+
+3. Monitor Ralph's progress in `.ralph/state.json`
+
+4. On completion, use generated pipeline from `.ralph/iterations/pipeline.md`
+
+### Manual Ralph Invocation
+
+For explicit Ralph mode on any pipeline:
+```bash
+export PIPELINE_USE_RALPH=true
+# or
+export RALPH_AUTO=true
+```
+
+Then invoke `/agent-pipeline-builder` as normal.
+
+### Ralph-Generated Pipeline Structure
+
+When Ralph builds the pipeline autonomously, it creates:
+
+```
+.claude/agents/[pipeline-name]/
+├── researcher.md    # Agent definition
+├── analyzer.md      # Agent definition
+└── writer.md        # Agent definition
+
+scripts/
+└── run-[pipeline-name].ts  # Orchestration script
+
+.ralph/
+├── PIPELINE.md      # Manifest
+├── state.json       # Progress tracking
+└── iterations/
+    └── pipeline.md  # Final generated pipeline
+```
+
+## Creating a Pipeline
+
+### Step 1: Define Pipeline Manifest
+
+Create a `pipeline.md` file:
+
+```markdown
+# Pipeline: [Name]
+
+## Stages
+1. researcher - Finds/fetches raw data
+2. analyzer - Processes and selects
+3. writer - Creates final output
+
+## Data Format
+All stages use JSON with markers: `<<<stage-name>>>...<<<end-stage-name>>>`
+```
+
+### Step 2: Create Agent Definitions
+
+For each stage, create an agent file `.claude/agents/[pipeline-name]/[stage-name].md`:
+
+```markdown
+---
+name: researcher
+description: What this agent does
+model: haiku  # or sonnet, opus
+---
+
+You are a [role] agent.
+
+## CRITICAL: NO EXPLANATION - JUST ACTION
+
+DO NOT explain what you will do. Just USE tools immediately, then output.
+
+## Instructions
+
+1. Use [specific tool] to get data
+2. Output JSON in the exact format below
+3. Wrap in markers as specified
+
+## Output Format
+
+<<<researcher>>>
+```json
+{
+  "data": [...]
+}
+```
+<<<end-researcher>>>
+```
+
+### Step 3: Implement Pipeline Script
+
+Create a script that orchestrates the agents:
+
+```typescript
+// scripts/run-pipeline.ts
+import { runAgent } from '@anthropic-ai/claude-agent-sdk';
+
+async function runPipeline() {
+  // Stage 1: Researcher
+  const research = await runAgent('researcher', {
+    context: { topic: 'AI news' }
+  });
+
+  // Stage 2: Analyzer (uses research output)
+  const analysis = await runAgent('analyzer', {
+    input: research,
+    context: { criteria: 'impact' }
+  });
+
+  // Stage 3: Writer (uses analysis output)
+  const final = await runAgent('writer', {
+    input: analysis,
+    context: { format: 'tweet' }
+  });
+
+  return final;
+}
+```
+
+## Pipeline Best Practices
+
+### 1. Single Responsibility
+Each agent does ONE thing:
+- ✓ researcher: Fetches data
+- ✓ analyzer: Filters and ranks
+- ✗ researcher-analyzer: Does both (too complex)
+
+### 2. Structured Data Flow
+- Use JSON or YAML for all inter-agent communication
+- Define schemas upfront
+- Validate output before passing to next stage
+
+### 3. Error Handling
+- Each agent should fail gracefully
+- Use fallback outputs
+- Log errors for debugging
+
+### 4. Deterministic Patterns
+- Constrain agents with specific tools
+- Use detailed system prompts
+- Avoid open-ended requests
+
+## Example Pipeline: AI News Tweet
+
+### Manifest
+```yaml
+name: ai-news-tweet
+stages:
+  - researcher: Gets today's AI news
+  - analyzer: Picks most impactful story
+  - writer: Crafts engaging tweet
+```
+
+### Researcher Agent
+```markdown
+---
+name: researcher
+description: Finds recent AI news using WebSearch
+model: haiku
+---
+
+Use WebSearch to find AI news from TODAY ONLY.
+
+Output:
+<<<researcher>>>
+```json
+{
+  "items": [
+    {
+      "title": "...",
+      "summary": "...",
+      "url": "...",
+      "published_at": "YYYY-MM-DD"
+    }
+  ]
+}
+```
+<<<end-researcher>>>
+```
+
+### Analyzer Agent
+```markdown
+---
+name: analyzer
+description: Analyzes news and selects best story
+model: sonnet
+---
+
+Input: Researcher output (stdin)
+
+Select the most impactful story based on:
+- Technical significance
+- Broad interest
+- Credibility of source
+
+Output:
+<<<analyzer>>>
+```json
+{
+  "selected": {
+    "title": "...",
+    "summary": "...",
+    "reasoning": "..."
+  }
+}
+```
+<<<end-analyzer>>>
+```
+
+### Writer Agent
+```markdown
+---
+name: writer
+description: Writes engaging tweet
+model: sonnet
+---
+
+Input: Analyzer output (stdin)
+
+Write a tweet that:
+- Hooks attention
+- Conveys key insight
+- Fits 280 characters
+- Includes relevant hashtags
+
+Output:
+<<<writer>>>
+```json
+{
+  "tweet": "...",
+  "hashtags": ["..."]
+}
+```
+<<<end-writer>>>
+```
+
+## Running the Pipeline
+
+### Method 1: Sequential Script
+```bash
+./scripts/run-pipeline.ts
+```
+
+### Method 2: Using Task Tool
+```typescript
+// Launch each stage as a separate agent task
+await Task('Research stage', researchPrompt, 'haiku');
+await Task('Analysis stage', analysisPrompt, 'sonnet');
+await Task('Writing stage', writingPrompt, 'sonnet');
+```
+
+### Method 3: Using Claude Code Skills
+Create a skill that orchestrates the pipeline with proper error handling.
+
+## Testing Pipelines
+
+### Unit Tests
+Test each agent independently:
+```bash
+# Test researcher
+npm run test:researcher
+
+# Test analyzer with mock data
+npm run test:analyzer
+
+# Test writer with mock analysis
+npm run test:writer
+```
+
+### Integration Tests
+Test full pipeline:
+```bash
+npm run test:pipeline
+```
+
+## Debugging Tips
+
+1. **Enable verbose logging** - See what each agent outputs
+2. **Validate JSON schemas** - Catch malformed data early
+3. **Use mock inputs** - Test downstream agents independently
+4. **Check marker format** - Agents must use exact markers
+
+## Common Patterns
+
+### Funnel Pattern
+```
+Many inputs → Filter → Select One → Output
+```
+Example: News aggregator → analyzer → best story
+
+### Transformation Pattern
+```
+Input → Transform → Validate → Output
+```
+Example: Raw data → clean → validate → structured data
+
+### Assembly Pattern
+```
+Part A + Part B → Assemble → Complete
+```
+Example: Research + style guide → formatted article
--- a/agent-pipeline-builder/ralph-pipeline.py
+++ b/agent-pipeline-builder/ralph-pipeline.py
@@ -0,0 +1,350 @@
+#!/usr/bin/env python3
+"""
+Ralph Integration for Agent Pipeline Builder
+
+Generates pipeline manifests for Ralph Orchestrator to autonomously build and test multi-agent pipelines.
+"""
+
+import os
+import sys
+import json
+import subprocess
+from pathlib import Path
+from typing import Optional, Dict, Any, List
+
+# Configuration
+RALPHLOOP_CMD = Path(__file__).parent.parent.parent.parent / "obsidian-web-interface" / "bin" / "ralphloop"
+PIPELINE_THRESHOLD = 3  # Minimum number of stages to trigger Ralph
+
+
+def analyze_pipeline_complexity(stages: List[Dict[str, str]]) -> int:
+    """
+    Analyze pipeline complexity and return estimated difficulty.
+
+    Returns: 1-10 scale
+    """
+    complexity = len(stages)  # Base: one point per stage
+
+    # Check for complex patterns
+    for stage in stages:
+        description = stage.get("description", "").lower()
+
+        # External data sources (+1)
+        if any(word in description for word in ["fetch", "api", "database", "web", "search"]):
+            complexity += 1
+
+        # Complex processing (+1)
+        if any(word in description for word in ["analyze", "transform", "aggregate", "compute"]):
+            complexity += 1
+
+        # Conditional logic (+1)
+        if any(word in description for word in ["filter", "validate", "check", "select"]):
+            complexity += 1
+
+    # Parallel stages add complexity
+    stage_names = [s.get("name", "") for s in stages]
+    if "parallel" in str(stage_names).lower():
+        complexity += 2
+
+    return min(10, complexity)
+
+
+def create_pipeline_manifest(stages: List[Dict[str, str]], manifest_path: str = ".ralph/PIPELINE.md") -> str:
+    """
+    Create a Ralph-formatted pipeline manifest.
+
+    Returns the path to the created manifest file.
+    """
+    ralph_dir = Path(".ralph")
+    ralph_dir.mkdir(exist_ok=True)
+
+    manifest_file = ralph_dir / "PIPELINE.md"
+
+    # Format the pipeline for Ralph
+    manifest_content = f"""# Pipeline: Multi-Agent Workflow
+
+## Stages
+
+"""
+    for i, stage in enumerate(stages, 1):
+        manifest_content += f"{i}. **{stage['name']}** - {stage['description']}\n"
+
+    manifest_content += f"""
+## Data Format
+
+All stages use JSON with markers: `<<<stage-name>>>...<<<end-stage-name>>>`
+
+## Task
+
+Build a complete multi-agent pipeline with the following stages:
+
+"""
+    for stage in stages:
+        manifest_content += f"""
+### {stage['name']}
+
+**Purpose:** {stage['description']}
+
+**Agent Configuration:**
+- Model: {stage.get('model', 'sonnet')}
+- Allowed Tools: {', '.join(stage.get('tools', ['Read', 'Write', 'Bash']))}
+
+**Output Format:**
+<<<{stage['name']}>>>
+```json
+{{
+  "result": "...",
+  "metadata": {{...}}
+}}
+```
+<<<end-{stage['name']}>>>
+
+"""
+
+    manifest_content += """
+## Success Criteria
+
+The pipeline is complete when:
+- [ ] All agent definitions are created in `.claude/agents/`
+- [ ] Pipeline orchestration script is implemented
+- [ ] Each stage is tested independently
+- [ ] End-to-end pipeline test passes
+- [ ] Error handling is verified
+- [ ] Documentation is complete
+
+## Instructions
+
+1. Create agent definition files for each stage
+2. Implement the pipeline orchestration script
+3. Test each stage independently with mock data
+4. Run the full end-to-end pipeline
+5. Verify error handling and edge cases
+6. Document usage and testing procedures
+
+When complete, add <!-- COMPLETE --> marker to this file.
+Output the final pipeline to `.ralph/iterations/pipeline.md`.
+"""
+
+    manifest_file.write_text(manifest_content)
+
+    return str(manifest_file)
+
+
+def should_use_ralph(stages: List[Dict[str, str]]) -> bool:
+    """
+    Determine if pipeline is complex enough to warrant RalphLoop.
+    """
+    # Check for explicit opt-in via environment
+    if os.getenv("RALPH_AUTO", "").lower() in ("true", "1", "yes"):
+        return True
+
+    if os.getenv("PIPELINE_USE_RALPH", "").lower() in ("true", "1", "yes"):
+        return True
+
+    # Check stage count
+    if len(stages) >= PIPELINE_THRESHOLD:
+        return True
+
+    # Check complexity
+    complexity = analyze_pipeline_complexity(stages)
+    return complexity >= 5
+
+
+def run_ralphloop_for_pipeline(stages: List[Dict[str, str]],
+                               pipeline_name: str = "multi-agent-pipeline",
+                               max_iterations: Optional[int] = None) -> Dict[str, Any]:
+    """
+    Run RalphLoop for autonomous pipeline construction.
+
+    Returns a dict with:
+    - success: bool
+    - iterations: int
+    - pipeline_path: str (path to generated pipeline)
+    - state: dict (Ralph's final state)
+    - error: str (if failed)
+    """
+    print("🔄 Delegating to RalphLoop 'Tackle Until Solved' for autonomous pipeline construction...")
+    print(f"   Stages: {len(stages)}")
+    print(f"   Complexity: {analyze_pipeline_complexity(stages)}/10")
+    print()
+
+    # Create pipeline manifest
+    manifest_path = create_pipeline_manifest(stages)
+    print(f"✅ Pipeline manifest created: {manifest_path}")
+    print()
+
+    # Check if ralphloop exists
+    if not RALPHLOOP_CMD.exists():
+        return {
+            "success": False,
+            "error": f"RalphLoop not found at {RALPHLOOP_CMD}",
+            "iterations": 0,
+            "pipeline_path": "",
+            "state": {}
+        }
+
+    # Build command - use the manifest file as input
+    cmd = [str(RALPHLOOP_CMD), "-i", manifest_path]
+
+    # Add optional parameters
+    if max_iterations:
+        cmd.extend(["--max-iterations", str(max_iterations)])
+
+    # Environment variables
+    env = os.environ.copy()
+    env.setdefault("RALPH_AGENT", "claude")
+    env.setdefault("RALPH_MAX_ITERATIONS", str(max_iterations or 100))
+
+    print(f"Command: {' '.join(cmd)}")
+    print("=" * 60)
+    print()
+
+    # Run RalphLoop
+    try:
+        process = subprocess.Popen(
+            cmd,
+            stdout=subprocess.PIPE,
+            stderr=subprocess.STDOUT,
+            text=True,
+            bufsize=1,
+            env=env
+        )
+
+        # Stream output
+        output_lines = []
+        for line in process.stdout:
+            print(line, end='', flush=True)
+            output_lines.append(line)
+
+        process.wait()
+        returncode = process.returncode
+
+        print()
+        print("=" * 60)
+
+        if returncode == 0:
+            # Read final state
+            state_file = Path(".ralph/state.json")
+            pipeline_file = Path(".ralph/iterations/pipeline.md")
+
+            state = {}
+            if state_file.exists():
+                state = json.loads(state_file.read_text())
+
+            pipeline_path = ""
+            if pipeline_file.exists():
+                pipeline_path = str(pipeline_file)
+
+            iterations = state.get("iteration", 0)
+
+            print(f"✅ Pipeline construction completed in {iterations} iterations")
+            if pipeline_path:
+                print(f"   Pipeline: {pipeline_path}")
+            print()
+
+            return {
+                "success": True,
+                "iterations": iterations,
+                "pipeline_path": pipeline_path,
+                "state": state,
+                "error": None
+            }
+        else:
+            return {
+                "success": False,
+                "error": f"RalphLoop exited with code {returncode}",
+                "iterations": 0,
+                "pipeline_path": "",
+                "state": {}
+            }
+
+    except KeyboardInterrupt:
+        print()
+        print("⚠️  RalphLoop interrupted by user")
+        return {
+            "success": False,
+            "error": "Interrupted by user",
+            "iterations": 0,
+            "pipeline_path": "",
+            "state": {}
+        }
+    except Exception as e:
+        return {
+            "success": False,
+            "error": str(e),
+            "iterations": 0,
+            "pipeline_path": "",
+            "state": {}
+        }
+
+
+def delegate_pipeline_to_ralph(stages: List[Dict[str, str]],
+                               pipeline_name: str = "multi-agent-pipeline") -> Optional[str]:
+    """
+    Main entry point: Delegate pipeline construction to Ralph if complex.
+
+    If Ralph is used, returns the path to the generated pipeline.
+    If pipeline is simple, returns None (caller should build directly).
+    """
+    if not should_use_ralph(stages):
+        return None
+
+    result = run_ralphloop_for_pipeline(stages, pipeline_name)
+
+    if result["success"]:
+        return result.get("pipeline_path", "")
+    else:
+        print(f"❌ RalphLoop failed: {result.get('error', 'Unknown error')}")
+        print("Falling back to direct pipeline construction...")
+        return None
+
+
+# Example pipeline stages for testing
+EXAMPLE_PIPELINE = [
+    {
+        "name": "researcher",
+        "description": "Finds and fetches raw data from various sources",
+        "model": "haiku",
+        "tools": ["WebSearch", "WebFetch", "Read"]
+    },
+    {
+        "name": "analyzer",
+        "description": "Processes data and selects best options",
+        "model": "sonnet",
+        "tools": ["Read", "Write", "Bash"]
+    },
+    {
+        "name": "writer",
+        "description": "Creates final output from analyzed data",
+        "model": "sonnet",
+        "tools": ["Write", "Edit"]
+    }
+]
+
+
+if __name__ == "__main__":
+    import argparse
+
+    parser = argparse.ArgumentParser(description="Test Ralph pipeline integration")
+    parser.add_argument("--test-complexity", action="store_true", help="Only test complexity")
+    parser.add_argument("--force", action="store_true", help="Force Ralph mode")
+    parser.add_argument("--example", action="store_true", help="Run with example pipeline")
+
+    args = parser.parse_args()
+
+    if args.test_complexity:
+        complexity = analyze_pipeline_complexity(EXAMPLE_PIPELINE)
+        print(f"Pipeline complexity: {complexity}/10")
+        print(f"Should use Ralph: {should_use_ralph(EXAMPLE_PIPELINE)}")
+    elif args.example:
+        if args.force:
+            os.environ["PIPELINE_USE_RALPH"] = "true"
+
+        result = delegate_pipeline_to_ralph(EXAMPLE_PIPELINE, "example-pipeline")
+
+        if result:
+            print("\n" + "=" * 60)
+            print(f"PIPELINE GENERATED: {result}")
+            print("=" * 60)
+        else:
+            print("\nPipeline not complex enough for Ralph. Building directly...")
--- a/agent-pipeline-builder/scripts/validate-pipeline.ts
+++ b/agent-pipeline-builder/scripts/validate-pipeline.ts
@@ -0,0 +1,146 @@
+#!/usr/bin/env bun
+/**
+ * Agent Pipeline Validator
+ *
+ * Validates pipeline manifest and agent definitions
+ * Usage: ./validate-pipeline.ts [pipeline-name]
+ */
+
+import { readFileSync, existsSync } from 'fs';
+import { join } from 'path';
+
+interface PipelineManifest {
+  name: string;
+  stages: Array<{ name: string; description: string }>;
+  dataFormat?: string;
+}
+
+interface AgentDefinition {
+  name: string;
+  description: string;
+  model?: string;
+}
+
+function parseFrontmatter(content: string): { frontmatter: any; content: string } {
+  const match = content.match(/^---\n([\s\S]+?)\n---\n([\s\S]*)$/);
+  if (!match) {
+    return { frontmatter: {}, content };
+  }
+
+  const frontmatter: any = {};
+  const lines = match[1].split('\n');
+  for (const line of lines) {
+    const [key, ...valueParts] = line.split(':');
+    if (key && valueParts.length > 0) {
+      const value = valueParts.join(':').trim();
+      frontmatter[key.trim()] = value;
+    }
+  }
+
+  return { frontmatter, content: match[2] };
+}
+
+function validateAgentFile(agentPath: string): { valid: boolean; errors: string[] } {
+  const errors: string[] = [];
+
+  if (!existsSync(agentPath)) {
+    return { valid: false, errors: [`Agent file not found: ${agentPath}`] };
+  }
+
+  const content = readFileSync(agentPath, 'utf-8');
+  const { frontmatter } = parseFrontmatter(content);
+
+  // Check required fields
+  if (!frontmatter.name) {
+    errors.push(`Missing 'name' in frontmatter`);
+  }
+
+  if (!frontmatter.description) {
+    errors.push(`Missing 'description' in frontmatter`);
+  }
+
+  // Check for output markers
+  const markerPattern = /<<<(\w+)>>>/g;
+  const markers = content.match(markerPattern);
+  if (!markers || markers.length < 2) {
+    errors.push(`Missing output markers (expected <<<stage>>>...<<<end-stage>>>)`);
+  }
+
+  return { valid: errors.length === 0, errors };
+}
+
+function validatePipeline(pipelineName: string): void {
+  const basePath = join(process.cwd(), '.claude', 'agents', pipelineName);
+  const manifestPath = join(basePath, 'pipeline.md');
+
+  console.log(`\n🔍 Validating pipeline: ${pipelineName}\n`);
+
+  // Check if pipeline directory exists
+  if (!existsSync(basePath)) {
+    console.error(`❌ Pipeline directory not found: ${basePath}`);
+    process.exit(1);
+  }
+
+  // Load and validate manifest
+  let stages: string[] = [];
+  if (existsSync(manifestPath)) {
+    const manifestContent = readFileSync(manifestPath, 'utf-8');
+    const { frontmatter } = parseFrontmatter(manifestContent);
+    stages = frontmatter.stages?.map((s: any) => typeof s === 'string' ? s : s.name) || [];
+  }
+
+  // If no manifest, auto-detect agents
+  if (stages.length === 0) {
+    const { readdirSync } = require('fs');
+    const files = readdirSync(basePath).filter((f: string) => f.endsWith('.md') && f !== 'pipeline.md');
+    stages = files.map((f: string) => f.replace('.md', ''));
+  }
+
+  console.log(`📋 Stages: ${stages.join(' → ')}\n`);
+
+  // Validate each agent
+  let hasErrors = false;
+  for (const stage of stages) {
+    const agentPath = join(basePath, `${stage}.md`);
+    const { valid, errors } = validateAgentFile(agentPath);
+
+    if (valid) {
+      console.log(`  ✅ ${stage}`);
+    } else {
+      console.log(`  ❌ ${stage}`);
+      for (const error of errors) {
+        console.log(`     ${error}`);
+      }
+      hasErrors = true;
+    }
+  }
+
+  // Check for scripts
+  const scriptsPath = join(process.cwd(), 'scripts', `run-${pipelineName}.ts`);
+  if (existsSync(scriptsPath)) {
+    console.log(`\n  ✅ Pipeline script: ${scriptsPath}`);
+  } else {
+    console.log(`\n  ⚠️  Missing pipeline script: ${scriptsPath}`);
+    console.log(`     Create this script to orchestrate the agents.`);
+  }
+
+  console.log('');
+
+  if (hasErrors) {
+    console.log('❌ Pipeline validation failed\n');
+    process.exit(1);
+  } else {
+    console.log('✅ Pipeline validation passed!\n');
+  }
+}
+
+// Main
+const pipelineName = process.argv[2];
+
+if (!pipelineName) {
+  console.log('Usage: validate-pipeline.ts <pipeline-name>');
+  console.log('Example: validate-pipeline.ts ai-news-tweet');
+  process.exit(1);
+}
+
+validatePipeline(pipelineName);