Awesome AI Agents › Reasoning Frameworks

anysphere/priompt

⭐ 2853 TypeScript added to this list on 2025-04-19 repository created 2023-06-23

Priompt is a JSX-based prompting library designed to optimize prompt construction for language models by using a priority system to decide what content to include within the context window. Inspired by web design libraries like React, Priompt allows developers to build prompts as JSX components, where each child element is assigned a priority that determines its inclusion based on token limits. This approach helps manage the limited token space effectively by including the most important parts of the prompt first and progressively adding less critical information as space allows. The library introduces several key components and concepts to facilitate prompt design. The component allows setting absolute or relative priorities for prompt sections, while ensures only the highest priority child is included, useful for fallback messages. Other components like reserve token space, parses output within the prompt, and enforces token limits on prompt sections to maintain consistency and caching benefits. Additional components such as , , and help structure message-based prompts, and supports image inclusion. Priompt also offers advanced features like callbacks for inclusion or exclusion of prompt parts, sourcemaps for mapping prompt text back to JSX components, and configuration options for stop tokens and response length. The library is designed to handle up to around 10,000 scopes efficiently, with plans for future enhancements including max token limits per scope and performance optimizations potentially involving Rust. While powerful, Priompt requires careful use of priorities to avoid caching inefficiencies and performance issues. It is open-source under the MIT license and welcomes contributions. Overall, Priompt provides a structured, flexible, and priority-driven approach to prompt design, enabling developers to create dynamic, token-efficient prompts for large language models.

https://github.com/anysphere/priompt

cachingcallbacksconfigurationcontext-windowfallback-messagesimage-inclusionjsxlanguage-modelsmessage-based-promptsmit-licenseopen-sourceoutput-captureperformance-optimizationpriority-systemprompt-designreact-inspiredscopesourcemapstoken-isolationtoken-limitstoken-reservation

Also in Reasoning Frameworks

bytedance/deer-flow

DeerFlow is a community-driven deep research framework that integrates language models with web search, crawling, and Python execution tools to facilitate comprehensive AI-powered research workflows.

VectifyAI/PageIndex

PageIndex is a reasoning-based document indexing and retrieval system that uses a hierarchical tree structure to enable human-like, vectorless retrieval and multi-step reasoning over long professional documents.

guidance-ai/guidance

Guidance is a programming paradigm and library that enables efficient and fine-grained control over large language models to produce structured, high-quality outputs with reduced latency and cost.

friuns2/BlackFriday-GPTs-Prompts

A curated repository of free GPT prompts and jailbreaks across various domains, enabling users to utilize GPT models without a subscription to GPT Plus.

refly-ai/refly

Refly is an open-source AI-native creation engine that combines multi-threaded dialogues, multimodal inputs, knowledge base integration, and AI-powered tools to transform ideas into production-ready interactive content and web applications.

mongodb-developer/GenAI-Showcase

MongoDB's GenAI Showcase is a comprehensive repository offering examples and applications that integrate MongoDB with advanced Generative AI techniques like Retrieval-Augmented Generation and AI Agents to support AI development and innovation.

HKUDS/Auto-Deep-Research

Auto-Deep-Research is an open-source, fully automated, and cost-effective personal AI assistant that supports multiple large language models and offers a one-click launch experience as an alternative to OpenAI's Deep Research.

NirDiamant/Controllable-RAG-Agent

Controllable-RAG-Agent is an advanced Retrieval-Augmented Generation solution featuring a deterministic graph-based autonomous agent designed for complex question answering with multi-step reasoning and hallucination prevention.