Agent frameworks

Napaxi is a mobile-native SDK for embedding on-device AI agent experiences directly into an app, giving Android, iOS and Flutter teams a shared Rust agent runtime with sandboxed tool execution while the host app keeps its own UI, accounts and model configuration.

• Agent frameworks
• Mobile

MASB is a FastAPI application that orchestrates a multi-agent system to answer questions by combining live web search with a FAISS-backed retrieval-augmented knowledge base, streaming its answer back over server-sent events.

• Agent frameworks
• RAG

Macaw is a private, on-device AI agent for macOS built around a small fine-tuned model that can call ninety-seven verified tools to control mail, files, calendar, music and system settings from a floating prompt bar, with no cloud calls or accounts.

• Agent frameworks

Invarail is a local, self-hosted authority layer for AI agents that separates what an agent is allowed to do from what it can learn to do, sitting in front of interchangeable workers for coding, repeatable procedures and open-ended tasks.

• Agent frameworks
• Security

Crow is a local AI agent rather than a chat box: it gives a model around thirty tools, MCP access, memory, skills, goals, subagents, vision and image generation and editing, running on one GPU through its own engine or llama.cpp, or against any OpenRouter model.

• Agent frameworks
• Inference

Android Agent Harness is a provider-neutral Kotlin SDK for building bounded AI agents on Android and the JVM, with a governed run kernel, budgeted context, an approval protocol for risky actions and a reference sample app.

• Agent frameworks
• Mobile

AgentClaw is a personal AI commander that understands intent, plans multi-step tasks and dispatches them to tools and skills, reachable through a web UI or through Telegram, WhatsApp, DingTalk, Feishu, QQ and WeCom, and doubling as a platform for hosting other independent agents.

• Agent frameworks

A long-horizon agent harness that decomposes an open-ended, multi-hour request into bounded subtasks, verifies each deliverable against the original goal, and compresses execution state into reusable memory.

• Agent frameworks

A Rust framework that builds an AI agent from a single declarative YAML specification, covering state machines, layered tool policy, twelve LLM providers with fallback, and a memory stack, with no code required for common cases.

• Agent frameworks

A lightweight CLI that lets AI agent runtimes such as Claude and Codex exchange messages through a shared, file-backed communication directory, with dry-run setup and no central daemon required.

• Agent frameworks

A single-binary, Go implementation of a RAG workspace and autonomous agent framework, inspired by the Leann backend, that runs embeddings and inference locally through Ollama or llama.cpp with AST-aware code indexing.

• Agent frameworks
• RAG

An LLM-powered tool that turns a natural-language causal question and a dataset into a full causal-inference analysis, selecting a suitable method, running it, checking diagnostics and explaining the result in plain language.

• Agent frameworks
• Data engineering

An autonomous agentic system that develops supervised machine-learning models for biomedical data, running experiments in a sandboxed Docker container and producing a trained model with a detailed PDF report.

• Agent frameworks
• Models

An open-source, local-first personal AI that remembers a user's goals and context across sessions and channels, and takes action on connected tools with confirmation required before high-impact steps.

• Agent frameworks

A TypeScript framework for building tool-using AI agents, from a single scripted agent to a multi-agent swarm, with a scaffolding CLI, an interactive setup wizard, and support for more than ten LLM providers.

• Agent frameworks

A community-curated list of frameworks, tools, libraries and platforms for building and deploying AI agents, covering coding, research, creative, web and voice agents plus memory, evaluation and deployment tooling.

• Agent frameworks
• Curated lists

A cryptographic trust layer for AI agent actions that signs every tool call as a hash-chained, offline-verifiable receipt, so an agent's actions and the authorization behind them can be proven rather than only logged.

• Agent frameworks
• Security

A small Ruby AI agent framework with an async message bus, multiple LLM providers, CLI, Telegram and WhatsApp channels, scheduling, browser automation, and separate memory personas per workspace.

• Agent frameworks

A benchmarking framework for single- and multi-agent LLM systems that pairs built-in datasets with evaluators, visual debugging of tool use and accuracy, and automated workflow optimization.

• Agent frameworks
• AI evals

A multi-provider Go library, split into about fifty per-vendor modules, that covers chat, embeddings, image generation, speech and an agent framework with handoffs, memory and Model Context Protocol tooling.

• Agent frameworks
• Libraries

A monorepo of Agent2Agent protocol wrappers that turn AI coding backends such as GitHub Copilot SDK, OpenCode, Claude Code, OpenAI Codex and Google Antigravity into standalone, spec-compliant A2A servers from a JSON configuration file.

• Coding agents
• Agent frameworks

Research code and the WildGUI dataset for a pipeline that extracts grounded GUI interaction trajectories from unlabeled internet videos to pretrain generalized graphical interface agents.

• Agent frameworks
• Papers

A distributed swarm of autonomous AI agents that turn open Lean 4 proof obligations into kernel-verified proofs, using the git repository as a work queue and continuous integration as the verification gate.

• Agent frameworks
• Testing

A research framework for measuring emergent safety failures that appear only when many AI agents interact, even when each individual agent behaves safely on its own.

• Agent frameworks
• AI evals