Tactus
Learn MoreGet StartedDownload

Examples / Human-in-the-Loop (HITL)

Human-in-the-Loop (HITL)

Patterns for integrating human decision-making into AI workflows. These examples show how to request human input, approval, and feedback during procedure execution.

Examples in This Chapter

Basic Hitl

Has Specs

Introduction to HITL primitives with direct procedure calls. This example demonstrates the traditional approach where the procedure code explicitly calls HITL primitives with hardcoded messages: - `Human.approve()` - Yes/no decisions - `Human.input()` - Text input with optional dropdown options - `Human.review()` - Content review and feedback

View example →

Agent Driven Approval

Has SpecsRequires API Keys

An AI agent that dynamically decides when to request human approval and constructs context-aware approval messages. This demonstrates **agent-driven HITL** where the agent has tools that wrap HITL primitives and uses them intelligently based on context.

View example →

← Previous ChapterClassificationNext Chapter →Standard Library

Want to contribute?

The examples repository is open source. Add your own examples or improve existing ones.

View on GitHub

Tactus is free and open source.

View Source
Tactus icon
Code Responsibly
Designed cybernetically
by Ryan Porter