Recipe

Recipe management for reusable agent configurations

The recipe command manages reusable agent recipes.

Usage

praisonai recipe [OPTIONS] COMMAND [ARGS]...

Commands

Command	Description
list	List available recipes
run	Run a recipe
create	Create recipe from natural language goal
optimize	Optimize existing recipe with AI feedback
init	Initialize a new recipe project
judge	Judge a trace with LLM
apply	Apply fixes from a judge plan

Run Options

Option	Description
--input, -i	Input JSON or file path
--config, -c	Config JSON overrides
--session, -s	Session ID for state grouping
--output, -o	Output mode: silent, status, trace, verbose, debug, json
--json	Output JSON (for parsing)
--stream	Stream output events (SSE-like)
--dry-run	Validate without executing
--explain	Show execution plan
--verbose, -v	Alias for --output verbose
--timeout <sec>	Timeout in seconds (default: 300)

Output Modes

Mode	Description
silent	No output (default, best performance)
status	Shows tool calls inline: ▸ tool → result ✓
trace	Timestamped execution trace: [HH:MM:SS] ▸ tool → result [0.2s] ✓
verbose	Full interactive output with panels
debug	Trace + metrics (tokens, cost, model)
json	Machine-readable JSONL events

Examples

List recipes

praisonai recipe list

Run a recipe

praisonai recipe run my-recipe

Run with status output

praisonai recipe run my-recipe --output status

Run with trace output

praisonai recipe run my-recipe --output trace

Run with input

praisonai recipe run my-recipe --input '{"query": "Hello"}'

Create Recipe from Goal

Create a complete recipe from a natural language goal. The AI automatically:

• Generates agents.yaml with appropriate agents
• Selects relevant tools based on the goal
• Creates TEMPLATE.yaml metadata
• Runs optimization loop (3 iterations by default)

1. Run Create Command

   praisonai recipe create "Build a web scraper for news articles"
   
2. Recipe Generated

   Creates folder with agents.yaml, TEMPLATE.yaml, and tools.py

3. Auto-Optimization

   Runs 3 optimization iterations with AI judge feedback

Create Options

Option	Description
--output, -o	Output directory (default: current)
--no-optimize	Skip optimization loop
--iterations	Number of optimization iterations (default: 3)
--threshold	Score threshold to stop (default: 8.0)

Examples

Basic

praisonai recipe create "Research AI trends and summarize"

praisonai recipe create "Build a calculator" --no-optimize

praisonai recipe create "Analyze stock data" --iterations 5 --threshold 9.0

---

Optimize Existing Recipe

Improve an existing recipe using AI judge feedback. Runs the recipe, evaluates output, and applies improvements.

1. Run Optimize

   praisonai recipe optimize my-recipe
   
2. Target Specific Area

   praisonai recipe optimize my-recipe "improve error handling"
   

Optimize Options

Option	Description
--iterations	Max optimization iterations (default: 3)
--threshold	Score threshold to stop (default: 8.0)
--input, -i	Input data for recipe runs

---

See Also

• Recipes - Recipe details
• Recipe Registry - Recipe registry