feat(LLMCall): BREAKING CHANGE Add support for LLM Executor

Include fixes for HTTP Executor
Updated tests and snapshot schema
Updated schema.json
Updated documentation

Author: ibacalu

.gitignore

@@ -211,7 +211,7 @@ __marimo__/
 .hooks/
 .mcp.json
 .scratch/
-local/
+.workflows/
 ARCHITECTURE.md
 CLAUDE.md
 docs/

README.md

@@ -70,6 +70,7 @@ Restart your LLM client, and you're ready to go!
 
 - **DAG-Based Workflows**: Automatic dependency resolution and parallel execution
 - **🔐 Secrets Management**: Server-side credential handling with automatic redaction (v5.0.0+)
+- **🤖 LLM Integration**: Native LLM API calls with schema validation and retry logic
 - **Workflow Composition**: Reusable workflows via Workflow blocks
 - **Conditional Execution**: Boolean expressions for dynamic control flow
 - **Variable Resolution**: Five-namespace system (inputs, metadata, blocks, secrets, __internal__)
@@ -114,6 +115,47 @@ blocks:
 }
 ```
 
+### 🤖 LLM Integration
+
+Call LLM APIs directly from workflows with automatic retry logic and JSON schema validation:
+
+```yaml
+blocks:
+  - id: extract_data
+    type: LLMCall
+    inputs:
+      provider: openai
+      model: gpt-4o-mini
+      api_key: "{{secrets.OPENAI_API_KEY}}"
+      prompt: "Extract key information from: {{inputs.text}}"
+      response_schema:
+        type: object
+        required: [summary, confidence]
+        properties:
+          summary: {type: string}
+          confidence: {type: number, minimum: 0, maximum: 1}
+      max_retries: 3
+```
+
+**Supported Providers:**
+
+- **OpenAI** - API access with native Structured Outputs
+- **Anthropic** - API access for Claude models
+- **Gemini** - API access for Gemini models
+- **Ollama** - Local Ollama server API
+- **OpenAI-compatible** - LM Studio, vLLM, and other compatible endpoints
+
+**Key Features:**
+
+- ✅ **Native schema validation** - Send JSON Schema to API for guaranteed structure
+- ✅ **Automatic retry** - Exponential backoff with validation feedback loop
+- ✅ **Token tracking** - Monitor usage and costs
+- ✅ **Secrets integration** - Secure API key management
+
+**For Local CLI Tools:**
+
+To use local Claude CLI or Gemini CLI instead of APIs, use the `llm-process` workflow template which wraps CLI commands.
+
 ## What Can You Do With It?
 
 ### Built-in Workflows

generate_schema.py

@@ -0,0 +1,42 @@
+#!/usr/bin/env python3
+"""Generate schema.json from Pydantic models.
+
+This script regenerates the workflow schema JSON file that provides:
+- VS Code YAML autocomplete
+- Workflow validation
+- Documentation generation
+
+Usage:
+    python generate_schema.py
+"""
+
+import json
+from pathlib import Path
+
+from workflows_mcp.engine.executor_base import create_default_registry
+
+
+def main():
+    """Generate and save schema.json."""
+    print("Generating workflow schema from executor registry...")
+
+    # Create registry with all built-in executors
+    registry = create_default_registry()
+
+    # Generate complete schema
+    schema = registry.generate_workflow_schema()
+
+    # Save to schema.json in project root
+    schema_path = Path(__file__).parent / "schema.json"
+
+    with open(schema_path, "w") as f:
+        json.dump(schema, f, indent=2)
+        f.write("\n")  # Add trailing newline
+
+    print(f"✓ Schema generated: {schema_path}")
+    print(f"  Executors: {len([k for k in schema.get('$defs', {}).keys() if k.endswith('Input')])}")
+    print(f"  Schema size: {schema_path.stat().st_size:,} bytes")
+
+
+if __name__ == "__main__":
+    main()