Reading

hitl: A Unified Human-in-the-Loop Approval SDK for AI Agents and Workflows

hitl is an open-source TypeScript SDK designed specifically for AI agents and persistent workflows. It provides a standardized Human-in-the-Loop approval mechanism, enabling automated processes to safely incorporate human oversight at critical decision points.

hitl人机协作Human-in-the-LoopAI智能体TypeScript工作流审批自动化持久化工作流SDK

Published 2026-06-14 22:46Recent activity 2026-06-14 22:51Estimated read 5 min

Section 01

Introduction: hitl — A Unified Human-in-the-Loop Approval SDK for AI Agents and Workflows

hitl is an open-source TypeScript SDK designed for AI agents and persistent workflows. It provides a standardized Human-in-the-Loop approval mechanism to bridge the trust gap in AI automation, incorporating human oversight at critical decision points to achieve a hybrid model of "machine autonomy + human supervision". The project is maintained by u17g, with code hosted on GitHub (link: https://github.com/u17g/hitl), and was released on June 14, 2026.

Section 02

Background: Trust Gap in AI Automation and the Need for Human Oversight

With the improvement of large language model capabilities, AI agents can perform complex tasks (writing code, calling APIs, managing databases, etc.), but autonomy brings risks. Full automation is costly in fields like financial transactions, medical diagnosis, and legal compliance, requiring the integration of human oversight. This is a key challenge for the implementation of AI applications.

Section 03

Methodology: Core Architecture and Solutions of hitl

hitl provides a unified human-machine collaboration framework, with the core idea of treating human-machine collaboration as reusable infrastructure. Its features include: 1. Persistent workflow support (event sourcing and state persistence to ensure reliable process recovery); 2. Clean TypeScript API (declarative definition of approval points without worrying about underlying details); 3. Flexible approval modes (single-person, multi-person joint signing, multi-level approval, conditional automatic approval).

Section 04

Evidence: Practical Application Scenarios of hitl

hitl can be applied in multiple scenarios: 1. Intelligent customer service: Transfer to humans during intense emotional situations or sensitive operations; 2. Code review: Request human review for sensitive operations after AI generates code; 3. Automated operation and maintenance: Request confirmation from operation and maintenance teams before configuration changes or fault recovery; 4. Content publishing: AI-generated content undergoes review by editors and legal teams.

Section 05

Technical Implementation Highlights: Security and Compliance Assurance

The technical highlights of hitl include: 1. Native TypeScript provides type safety, catching errors at compile time; 2. Extensible notification system (email, Slack, Webhook, etc.); 3. Audit and compliance: All approval activities are persistently recorded (request time, approver, decision result, etc.) to meet industry compliance requirements.

Section 06

Open Source Ecosystem and Future Outlook

hitl is open-source and hosted on GitHub, using a permissive license that allows commercial use, and provides detailed documentation and examples. In the future, human-machine collaboration will become a standard practice for AI applications. hitl promotes the transition of AI applications from prototype verification to engineering implementation, and we look forward to more AI security tools emerging.

Continue Reading

Keep going with more reads from the same topic.

Nornir MCP Server: An Enterprise-Grade Bridge for Integrating Large Language Models into Network Automation

Nornir MCP Server is an enterprise-level server based on the Model Context Protocol (MCP). It seamlessly integrates large language models (such as Claude) with the Nornir network automation framework, supporting natural language orchestration for multi-vendor network devices (Cisco, Arista, Juniper, etc.), and providing production-grade features like a dual-engine architecture (NAPALM + Netmiko), intelligent filtering, and a secure sandbox.

Recent activity 2026-05-06 20:51

Bibliothèque Française LLM: A French Public Domain Literature Index System Optimized for Large Language Models

Bibliothèque Française LLM is a structured indexing and annotation project for French public domain literature designed specifically for large language models (LLMs). It integrates multiple authoritative sources such as DraCor, Common Corpus, and Wikisource, providing metadata indexing categorized by genre, author, and era, as well as in-depth annotations for dramatic texts (including characters, lines, stage directions, etc.). Its aim is to enable LLMs to efficiently read and understand classic French literary works.

Recent activity 2026-05-06 20:50

Splinter: A Lock-Free Zero-Copy Shared Memory KV and Vector Storage Library That Eliminates Socket and Memcpy Overhead for LLM Inference

Splinter is a minimalist, high-performance key-value (KV) and vector storage system enabling zero-latency inter-process communication via shared memory and atomic operations. With only 766 lines of core code, it supports millions of operations per second and 768-dimensional vector storage, offering a new architectural approach for local LLM inference and data-intensive applications.

Recent activity 2026-04-03 08:49

libmlxforge: An Embedded MLX LLM Inference Engine for Apple Silicon

libmlxforge is an embeddable MLX large language model (LLM) inference engine designed specifically for Apple Silicon. It provides a unified C ABI interface, supports calls from Node.js, Swift, and Rust, and features continuous batching, streaming output, JSON-constrained structured output, and embedding vector generation.

Recent activity 2026-06-09 17:23