demo
Demo Skill
Demonstrates UltraRAG-style execution tracing with native workflows + YAML pipelines.
Tools
hello
Simple demo command for testing hot reload.
Parameters:
name(string, optional): Name to greet (default: "Guest")
echo
Echo back the input message.
Parameters:
message(string, optional): Message to echo (default: "Hello!")
test_yaml_pipeline
Test YAML pipeline compilation and execution with omni.tracer.
Parameters:
pipeline_type(string, optional): Type of pipeline to test."simple": Sequential pipeline (analyze → draft → finalize)"loop": Pipeline with iteration loop (analyze → evaluate → reflect)"branch": Pipeline with conditional branching"rag": Full RAG pipeline with retrieval
run_graphflow
Execute packaged graphflow runtime with trace output and quality gates.
Parameters:
scenario(string, optional): Scenario name ("simple"or"complex")quality_threshold(string, optional): Draft/finalize quality thresholdquality_gate_novelty_threshold(string, optional): Gate novelty thresholdquality_gate_coverage_threshold(string, optional): Gate coverage thresholdquality_gate_min_evidence_count(string, optional): Minimum evidence countquality_gate_require_tradeoff(string, optional): Require trade-off in outputquality_gate_max_fail_streak(string, optional): Max gate-fail streak before force-draft
list_pipeline_examples
List available YAML pipeline test examples.
Returns:
- Available pipeline types and their descriptions
- Usage instructions
Pipeline Examples
The demo skill includes YAML pipeline files in pipelines/ directory:
| File | Description | Features |
|---|---|---|
simple.yaml |
Sequential pipeline | analyze → draft → finalize |
loop.yaml |
Iterative reflection | quality evaluation + loop |
branch.yaml |
Conditional routing | router-based branching |
rag.yaml |
Full RAG | retrieval + generation + evaluation |
YAML Pipeline Structure
servers:
generator: builtin
parameters:
topic: "$topic"
pipeline:
- generator.analyze
- generator.draft
- generator.finalize
runtime:
checkpointer:
kind: memory
tracer:
callback_dispatch_mode: inline
UltraRAG Features
This demo showcases:
- YAML pipelines - Declarative workflow definitions
- Loop control - Iterative reflection with max_iterations
- Conditional branching - Router-based routing
- Memory tracking - Variables saved with history
- Rich output - Colored step-by-step tracing
Usage
# Simple greeting
omni skill run demo.hello --name "World"
# Echo test
omni skill run demo.echo --message "Test message"
# List available pipeline examples
omni skill run demo.list_pipeline_examples
# Test YAML pipelines
omni skill run demo.test_yaml_pipeline --pipeline_type "simple"
omni skill run demo.test_yaml_pipeline --pipeline_type "loop"
omni skill run demo.test_yaml_pipeline --pipeline_type "branch"
omni skill run demo.test_yaml_pipeline --pipeline_type "rag"
# Run packaged graphflow runtime
omni skill run demo.run_graphflow --scenario "complex"
More from tao3k/omni-dev-fusion
crawl4ai
Use when crawling web pages, extracting markdown content, or scraping website data with intelligent chunking and skeleton planning. Use when the user provides a URL or link to fetch or crawl.
50software_engineering
Use when analyzing project architecture, exploring codebase structure, understanding system design, reviewing code patterns, or navigating modular components.
18python_engineering
Use when linting Python code, formatting with ruff/black, running pytest tests, type checking with pyright, or modernizing Python 3.12+ standards.
18code_tools
Use when searching code by structure or meaning, analyzing code patterns, finding class or function definitions, or exploring codebase architecture.
16rust_engineering
Use when analyzing Rust project structure, managing Cargo dependencies, building and testing Rust projects, or generating Rust code.
16git
Use when committing code, managing branches, pushing to remote, creating pull requests, or performing version control operations. Conforms to docs/reference/skill-routing-value-standard.md.
15