Orcha
Tutorial and Examples

Agent with skills

Example OrchaJS agent that loads specialized skill instructions only when they are relevant.

agent-with-skills~/agent-with-skills/orcha/writingAgent/skills/technicalExplanation/index.json
01{02  "name": "technical_explanation",03  "description": "Explain a technical concept with a practical example.",04  "triggers": [05    "The user asks how a technical concept works.",06    "The user asks for a practical engineering example."07  ]08}

Focused knowledge, loaded on demand

This example is a complete technical writing agent with one skill for explaining engineering concepts. The base agent stays small. It initially sees only the skill's name, description, and triggers.

When a request matches, the model loads the full procedure through Orcha's internal load_skill tool. Orcha continues the same run automatically, and the loaded instructions remain active for the durable session.

A skill is the right fit for policy, process, domain knowledge, or a repeatable method. It guides the model but does not execute code.

Select any file in the editor to inspect the complete standalone project.

Understand each file

orcha/writingAgent/skills/technicalExplanation/index.json

Provides the small discovery card the model sees before loading the skill.

  • name is the model-facing skill identifier.
  • description says what the skill helps with.
  • triggers describe situations in which it should be loaded.
orcha/writingAgent/skills/technicalExplanation/instructions.md

Contains the complete procedure, loaded only when the model chooses this skill.

  • Loaded instructions remain active for the durable session.
  • A skill adds guidance; it does not execute application code.
orcha/writingAgent/index.json

Configures the writing agent and its model.

orcha/writingAgent/instructions.md

Defines the base behavior that is always present before any skill is loaded.

orcha/index.ts

Configures the provider and registers writingAgent. Skill folders are discovered automatically.

app.ts

Runs a request that matches the skill's triggers and prints the completed response.

Next: build an agent that takes action.