Hugin Agent Creation Guide
This skill helps you create Hugin AI agents. It provides an overview and directs you to detailed reference documentation.
How to Use This Skill
- Quick Start: Follow the minimal agent example below
- Detailed Info: Read the appropriate reference file for schemas and examples
- Scaffolding: Use
/hugin-agent-creator:hugin-scaffoldto generate starter files
Reference Files (Read These for Details)
When the user needs detailed information about a specific topic, read the appropriate reference file from this plugin directory:
| Topic | Reference File | When to Read |
|---|---|---|
| Config files | references/config-reference.md | User asks about config options, models, tools |
| Task files | references/task-reference.md | User asks about tasks, parameters, prompts, pipelines |
| System templates | references/template-reference.md | User asks about system prompts, agent personality |
| Custom tools | references/tool-reference.md | User wants to create tools, use ToolResponse/AgentCall/AskHuman |
| Agent patterns | references/patterns.md | User needs architecture guidance, pattern examples |
Important: These reference files contain detailed schemas, field descriptions, and complete examples. Read them when answering specific questions.
Starter Templates
When the user wants to create files, read and adapt the starter templates from this plugin:
| Template | File | Use For |
|---|---|---|
| Config | templates/minimal-config.yaml | Creating agent configs |
| Task | templates/minimal-task.yaml | Creating task definitions |
| System template | templates/minimal-template.yaml | Creating system prompts |
| Tool definition | templates/tool-definition.yaml | Creating tool YAML |
| Tool implementation | templates/tool-implementation.py | Creating tool Python code |
Quick Start: Minimal Agent (3 Files)
A working agent needs just 3 files:
my_agent/
├── configs/my_agent.yaml # Agent configuration
├── tasks/my_task.yaml # Task definition
└── templates/my_system.yaml # System prompt template
Run Command
uv run hugin run --task my_task --task-path ./my_agent
Core Concepts (Overview)
Config
Defines an agent's identity: model, system template, available tools.
- Read
references/config-reference.mdfor full schema
Task
Defines what an agent should do: prompt, parameters, optional pipeline.
- Read
references/task-reference.mdfor full schema
Template
System prompt that sets agent behavior and personality.
- Read
references/template-reference.mdfor examples
Tool
Extends agent capabilities with Python code.
- Read
references/tool-reference.mdfor implementation guide
Built-in Tools
Reference with builtins.<tool>:<alias>:
| Tool | Description |
|---|---|
builtins.finish:finish | Complete task with success/failure |
builtins.ask_user:ask_user | Ask user a question (requires interactive: true) |
builtins.save_insight:save_insight | Save findings as artifacts |
builtins.launch_agent:launch_agent | Spawn sub-agents |
builtins.list_agents:list_agents | List available agents |
builtins.read_file:read_file | Read file contents (safe, read-only) |
builtins.list_files:list_files | List directory contents with glob patterns |
builtins.search_files:search_files | Search files for patterns (grep-like) |
Decision Tree: Which Pattern?
Do you need custom tools?
├── No → Minimal pattern
└── Yes → Tool pattern
└── Does the tool spawn another agent?
├── No → Simple tool
└── Yes → AgentCall pattern
Do you need human input during execution?
├── No → Set interactive: false
└── Yes → AskHuman pattern (interactive: true)
Do you need multiple processing stages?
├── No → Single task
└── Yes → Task sequence pattern (task_sequence + pass_result_as)
Do you need multiple agents running together?
└── Yes → Multi-agent pattern (shared state via env_vars)
For detailed pattern examples, read references/patterns.md
Common Patterns (Summary)
- Minimal Agent - Config + task + template, built-in tools only
- Tool Agent - Custom tools for specific capabilities
- Pipeline Agent - Multi-stage processing with result passing
- Human-in-the-Loop - Requires human approval during execution
- Agent Delegation - Spawns specialized sub-agents
- Multi-Agent - Multiple agents sharing state
Workflow for Helping Users
- Understand what they want to build - Ask clarifying questions if needed
- Recommend a pattern - Use the decision tree above
- Read the relevant reference - Get detailed schema information
- Read the starter template - Use as base for their files
- Customize for their use case - Replace placeholders, add their logic
- Provide the run command - Show how to test the agent