nexu-io/open-design
Open Design is a local-first, open-source desktop application that serves as an alternative to Claude Design, integrating with various coding agents to generate web, desktop, and mobile prototypes,...
Awesome AI Agents › Generative Software Frameworks
Agent-Field/SWE-AF
SWE-AF is an autonomous software engineering runtime built on the AgentField platform. A single call to the swe-planner build endpoint, made either through the af command-line tool or by posting JSON to the local HTTP API, starts a fleet of role-specialised agents that behave as an engineering team: product managers scope the goal, architects design, coders implement, reviewers review and testers verify, with the result delivered as production-grade pull requests. The input is a natural-language goal plus a repository, and a configuration block that selects the underlying coding runtime, for example Claude Code, and assigns a model per role so that cheaper models handle routine steps while stronger models are reserved for coding and QA. Two operating modes are documented. Single-repository mode is the default and takes either a remote repository URL or a local path. Multi-repository mode takes an array of repositories with roles, where a primary repository drives the build and failures there block progress, while dependency repositories are libraries or services modified in support of the primary one and their failures are recorded without blocking. That mode targets work spanning a main application plus a shared SDK, monorepo sub-projects split across repositories, or a feature crossing several microservices. An optional learning flag lets runs accumulate reusable knowledge, and the project describes adaptive factory control for scaling from simple goals to programs involving hundreds or thousands of agent invocations. The README presents an autonomous build case study on a Rust-based Python interpreter runtime with throughput comparisons, together with a methodology note explaining that the measured baseline is subprocess spawning rather than an in-process interpreter. The project is in public beta, requires Python 3.12 or newer and a recent af release, and is Apache 2.0 licensed.
https://github.com/Agent-Field/SWE-AF
Also in Generative Software Frameworks
deepset-ai/haystack
Haystack is an AI orchestration framework for building customizable, production-ready NLP applications using large language models, vector search, and advanced retrieval techniques.
op7418/guizang-ppt-skill
AI agent skill for generating polished HTML slide decks with various design styles, image prompts, social covers, and a WebGL/low-power presentation runtime.
HKUDS/Agentic-AIGC
ViMax is an agentic AI system for end-to-end video generation, transforming raw ideas, novels, or scripts into complete video content by orchestrating scriptwriting, storyboarding, character creati...
The-Pocket/PocketFlow
Pocket Flow is a minimalist 100-line LLM framework that uses a graph-based abstraction to enable expressive multi-agent systems, workflows, and retrieval-augmented generation, designed for high productivity and minimal bloat.
Netflix/metaflow
Metaflow is a human-centric framework by Netflix that streamlines building, managing, and deploying real-life AI and machine learning systems from prototyping to production at scale.
presenton/presenton
Presenton is an open-source AI-powered presentation generator and API that runs locally, offering customizable layouts, multiple AI model integrations, versatile image generation, and privacy-focused presentation creation and export.
thesysdev/openui
OpenUI is a full-stack Generative UI framework that offers a compact, streaming-first language for model-generated UI, a React runtime, and token-efficient chat interfaces.