diff --git a/README.md b/README.md index bd4f013..8cb1d41 100644 --- a/README.md +++ b/README.md @@ -1,7 +1,7 @@ # Awesome Agents [![Awesome](https://awesome.re/badge.svg)](https://awesome.re) - -[![BCOS Certified](https://img.shields.io/badge/BCOS-Certified-brightgreen?style=flat-square&logo=data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCI+PHBhdGggZmlsbD0id2hpdGUiIGQ9Ik05IDE2LjE3TDQuODMgMTJsLTEuNDIgMS40MUw5IDE5IDIxIDdsLTEuNDEtMS40MXoiLz48L3N2Zz4=)](https://github.com/Scottcjn/Rustchain/blob/main/BCOS.md) - + +[![BCOS Certified](https://img.shields.io/badge/BCOS-Certified-brightgreen?style=flat-square&logo=data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCI+PHBhdGggZmlsbD0id2hpdGUiIGQ9Ik05IDE2LjE3TDQuODMgMTJsLTEuNDIgMS40MUw5IDE5IDIxIDdsLTEuNDEtMS40MXoiLz48L3N2Zz4=)](https://github.com/Scottcjn/Rustchain/blob/main/BCOS.md) + > A curated list of AI agent platforms, frameworks, protocols, tools, and resources. @@ -49,8 +49,8 @@ AI agents are autonomous software entities that perceive their environment, make - [Letta](https://github.com/letta-ai/letta) - Framework for creating stateful LLM agents with long-term memory (formerly MemGPT). - [Bolt.new](https://bolt.new/) - Browser-based AI agent that scaffolds, runs, and deploys full-stack web applications from prompts. - [Onepilot](https://onepilotapp.com) - iOS and iPadOS SSH client that runs and orchestrates terminal coding agents (Claude Code, Codex CLI, OpenClaw, Hermes) on a remote machine. -- [Skales](https://skales.app) - Local-first desktop AI agent that runs goals autonomously in the background, with multi-agent teams, desktop and browser automation, and 15+ providers or fully offline via Ollama. -- [Hivekeep](https://github.com/MarlBurroW/hivekeep) - Self-hosted platform to run a team of specialized AI agents with persistent memory and a web UI, reachable over Telegram, Slack, Discord and Matrix, in a single Bun and SQLite container. +- [Skales](https://skales.app) - Local-first desktop AI agent that runs goals autonomously in the background, with multi-agent teams, desktop and browser automation, and 15+ providers or fully offline via Ollama. +- [Hivekeep](https://github.com/MarlBurroW/hivekeep) - Self-hosted platform to run a team of specialized AI agents with persistent memory and a web UI, reachable over Telegram, Slack, Discord and Matrix, in a single Bun and SQLite container. ## Frameworks @@ -76,7 +76,7 @@ AI agents are autonomous software entities that perceive their environment, make - [Camel](https://github.com/camel-ai/camel) - Communicative agents framework for exploring multi-agent cooperation and competition. - [Agency Swarm](https://github.com/VRSEN/agency-swarm) - Framework for creating collaborative swarms of AI agents based on the agency model. - [TaskWeaver](https://github.com/microsoft/TaskWeaver) - Microsoft's code-first agent framework converting natural language requests into executable code. -- [Mastra](https://github.com/mastra-ai/mastra) - TypeScript framework for building AI applications with agents, workflows, and RAG. +- [Mastra](https://github.com/mastra-ai/mastra) - TypeScript framework for building AI applications with agents, workflows, and RAG. - [rust-norion](https://github.com/yanghao1143/rust-norion) - Rust prototype for building agent runtime control layers with memory gates, model routing policy, audit evidence, and self-evolution loop tooling. - [Agno](https://github.com/agno-agi/agno) - Lightweight library for building multi-modal agents with memory and knowledge. - [Bernstein](https://github.com/sipyourdrink-ltd/bernstein) - Python orchestrator that drives 40+ CLI coding agents (Claude Code, Codex, Gemini CLI, Cursor, Aider) in parallel git worktrees with deterministic scheduling, quality gates, and an HMAC-chained audit log. @@ -91,7 +91,7 @@ AI agents are autonomous software entities that perceive their environment, make - [Grazer](https://github.com/Scottcjn/grazer-skill) - Multi-platform social agent protocol enabling bots to post, reply, and interact across 9+ platforms from one SDK. - [Agent Protocol](https://github.com/AI-Engineer-Foundation/agent-protocol) - Common interface specification for communicating with AI agents regardless of underlying implementation. - [Agora Protocol](https://github.com/agoraprotocol/agora) - Open protocol for agent-to-agent marketplace interactions and task negotiation. -- [LMOS](https://eclipse.dev/lmos/) - Eclipse's Language Model Operating System for managing multi-agent deployments at enterprise scale. +- [LMOS](https://eclipse.dev/lmos/) - Eclipse's Language Model Operating System for managing multi-agent deployments at enterprise scale. - [ANP](https://github.com/agent-network-protocol/AgentNetworkProtocol) - Agent Network Protocol for building an open, secure, and efficient collaboration network for AI agents. ## Video and Media @@ -123,7 +123,7 @@ AI agents are autonomous software entities that perceive their environment, make - [SwarmHub](https://swarmhub.ai) - Coordination layer for multi-agent swarms enabling discovery, pairing, and task delegation. - [Agent Directory](https://agentdirectory.ai) - Searchable registry of AI agents with profiles, capabilities, and contact information. - [Character.AI](https://character.ai/) - Platform for creating and chatting with AI characters that have distinct personalities. -- [Kissable](https://kissable.app) - AI companion with persistent memory, together photos, video generation, community scenario catalog, emotional voice, and unlimited messages. +- [Kissable](https://kissable.app) - AI companion with persistent memory, together photos, video generation, community scenario catalog, emotional voice, and unlimited messages. - [SillyTavern](https://github.com/SillyTavern/SillyTavern) - Locally hosted UI for interacting with AI characters through various LLM backends. - [Hugging Face Spaces](https://huggingface.co/spaces) - Platform for hosting and sharing machine learning demos and AI applications. @@ -135,26 +135,26 @@ AI agents are autonomous software entities that perceive their environment, make - [wRTC Token (Solana)](https://solscan.io/token/12TAdKXxcGf6oCv4rqDz2NkgxjyHq6HQKoxKZGf5i4X) - Wrapped RTC token bridged to Solana with a Raydium liquidity pool. - [wRTC Token (Base)](https://basescan.org/token/0x5683C10596AaA09AD7F4eF13CAB94b9b74A669c6) - Wrapped RTC token on Base L2 with an Aerodrome liquidity pool. - [x402 Payment Protocol](https://github.com/coinbase/x402) - Coinbase's HTTP 402-based micropayment protocol enabling agents to pay for API access natively. -- [x402-proxy](https://github.com/cascade-protocol/x402-proxy) - `curl` for x402 paid APIs. CLI that auto-pays HTTP 402 responses with USDC on Base and Solana, with MCP stdio proxy for AI agents. -- [TWZRD Agent Intel](https://intel.twzrd.xyz) - Trust scoring for AI agents on Solana. Verify agent wallet identity before x402 micropayments. Free MCP: `{"mcpServers":{"twzrd-agent-intel":{"url":"https://intel.twzrd.xyz/mcp"}}}` +- [x402-proxy](https://github.com/cascade-protocol/x402-proxy) - `curl` for x402 paid APIs. CLI that auto-pays HTTP 402 responses with USDC on Base and Solana, with MCP stdio proxy for AI agents. +- [TWZRD Agent Intel](https://intel.twzrd.xyz) - Trust scoring for AI agents on Solana. Verify agent wallet identity before x402 micropayments. Free MCP: `{"mcpServers":{"twzrd-agent-intel":{"url":"https://intel.twzrd.xyz/mcp"}}}` - [Fetch.ai](https://github.com/fetchai/fetchd) - Decentralized machine learning platform with autonomous economic agents on a dedicated blockchain. - [Autonolas](https://github.com/valory-xyz/open-autonomy) - Platform for creating and deploying autonomous agent services on-chain. - [Virtuals Protocol](https://www.virtuals.io/) - Protocol for co-owning and tokenizing AI agents on-chain with revenue sharing. -- [Hashgraph Online (HOL)](https://hol.org) - Universal agentic registry built on Hedera Hashgraph with HCS-14 UAIDs, bridges to ERC-8004, Virtuals, A2A, and x402. -- [AIMorgan](https://aimorgan.vercel.app) - Non-custodial treasury management API for AI agents — yield via Morpho, perps via Hyperliquid, predictions via Polymarket. x402-native with MCP server. +- [Hashgraph Online (HOL)](https://hol.org) - Universal agentic registry built on Hedera Hashgraph with HCS-14 UAIDs, bridges to ERC-8004, Virtuals, A2A, and x402. +- [AIMorgan](https://aimorgan.vercel.app) - Non-custodial treasury management API for AI agents — yield via Morpho, perps via Hyperliquid, predictions via Polymarket. x402-native with MCP server. - [ai16z ELIZA](https://github.com/ai16z/eliza) - Multi-agent simulation framework enabling AI agents to interact on social media and manage crypto wallets. - [Morpheus](https://github.com/MorpheusAIs/Morpheus) - Decentralized network of personal AI agents connected to smart contracts and crypto wallets. - [SingularityNET](https://github.com/singnet) - Decentralized marketplace for AI services where agents can buy and sell algorithms. -- [dPaPay](https://dpapay.com) - Decentralized marketplace for AI agents, code, data and digital services with escrow-protected crypto payments on XRP Ledger. List agents, prompts, workflows. Instant settlement, ~$350 free volume, no platform lock-in. +- [dPaPay](https://dpapay.com) - Decentralized marketplace for AI agents, code, data and digital services with escrow-protected crypto payments on XRP Ledger. List agents, prompts, workflows. Instant settlement, ~$350 free volume, no platform lock-in. - [Ocean Protocol](https://github.com/oceanprotocol) - Decentralized data exchange protocol enabling AI agents to access and monetize data. -- [AgentsCoin](https://agents-coin.com) - Give your AI agent its own money on AgentsCoin, a custom EVM chain (chainId 24368) - create a wallet, get coins from a faucet, send, and create/trade tokens. [MCP server](https://agents-coin.com/mcp) plus a one-click Claude Desktop extension. +- [AgentsCoin](https://agents-coin.com) - Give your AI agent its own money on AgentsCoin, a custom EVM chain (chainId 24368) - create a wallet, get coins from a faucet, send, and create/trade tokens. [MCP server](https://agents-coin.com/mcp) plus a one-click Claude Desktop extension. ## Monitoring and Observability *Tools for monitoring, debugging, and evaluating AI agent performance.* - [agenttrace](https://github.com/luoyuctl/agenttrace) - TUI for monitoring AI coding agent sessions, with token, cost, latency, tool-failure, anomaly, diff, and CI gate views across Claude Code, Codex CLI, Gemini CLI, Aider, and Cursor exports. -- [ax (agent telemetry)](https://github.com/Necmttn/ax) - Local-first telemetry graph for coding agents, with read-only MCP queries for sessions, tool calls, skills, and costs across Claude Code, Codex, Cursor, OpenCode, and Pi. +- [ax (agent telemetry)](https://github.com/Necmttn/ax) - Local-first telemetry graph for coding agents, with read-only MCP queries for sessions, tool calls, skills, and costs across Claude Code, Codex, Cursor, OpenCode, and Pi. - [OpenClaw Beacon Scorecard](https://github.com/Scottcjn/beacon-skill) - Liveness and capability scoring system for agents participating in the Beacon discovery network. - [Not Human Search](https://nothumansearch.ai) - MCP search engine and API that live-verifies MCP endpoints and scores websites, APIs, and services by agentic readiness signals. - [LangSmith](https://smith.langchain.com/) - LangChain's platform for debugging, testing, evaluating, and monitoring LLM applications and agents. @@ -166,7 +166,7 @@ AI agents are autonomous software entities that perceive their environment, make - [Weights & Biases](https://github.com/wandb/wandb) - Platform for experiment tracking, model management, and ML pipeline observability. - [Portkey](https://github.com/Portkey-AI/gateway) - AI gateway for routing, monitoring, and managing requests across 200+ LLM providers. - [AgentOps](https://github.com/AgentOps-AI/agentops) - Toolkit for agent monitoring, testing, and replay debugging with session recordings. -- [BrowserTrace](https://github.com/aaronlab/browsertrace) - Local-first trace viewer for debugging failed AI browser-agent and computer-use runs with screenshots, URLs, actions, model output, status, and redacted shareable exports. +- [BrowserTrace](https://github.com/aaronlab/browsertrace) - Local-first trace viewer for debugging failed AI browser-agent and computer-use runs with screenshots, URLs, actions, model output, status, and redacted shareable exports. ## SDKs and Libraries @@ -182,7 +182,7 @@ AI agents are autonomous software entities that perceive their environment, make - [Ollama Python](https://github.com/ollama/ollama-python) - Python library for running and interacting with local LLMs via the Ollama runtime. - [LiteLLM](https://github.com/BerriAI/litellm) - Unified interface to call 100+ LLM APIs using the OpenAI format with load balancing and spend tracking. - [Marvin](https://github.com/prefecthq/marvin) - Lightweight AI engineering toolkit for building natural language interfaces and AI functions. -- [AgentWallet SDK](https://github.com/up2itnow0822/agentwallet-sdk) - Non-custodial wallet SDK for AI agents with x402 payments, CCTP cross-chain bridging, and SpendingPolicy guardrails. +- [AgentWallet SDK](https://github.com/up2itnow0822/agentwallet-sdk) - Non-custodial wallet SDK for AI agents with x402 payments, CCTP cross-chain bridging, and SpendingPolicy guardrails. ## Standards and Specifications @@ -192,7 +192,7 @@ AI agents are autonomous software entities that perceive their environment, make - [OpenAPI Specification](https://github.com/OAI/OpenAPI-Specification) - Industry-standard specification for describing HTTP APIs, widely used for agent tool definitions. - [JSON-RPC 2.0](https://www.jsonrpc.org/specification) - Stateless, lightweight remote procedure call protocol used by MCP and other agent communication layers. - [A2A Protocol Spec](https://github.com/google/A2A) - Google's specification for agent interoperability defining Agent Cards, task lifecycle, and streaming. -- [MCP Specification](https://modelcontextprotocol.io/specification/) - Anthropic's specification for the Model Context Protocol defining tools, resources, and prompts. +- [MCP Specification](https://modelcontextprotocol.io/specification/) - Anthropic's specification for the Model Context Protocol defining tools, resources, and prompts. - [Tool Use Schemas](https://json-schema.org/) - JSON Schema, the foundation for defining tool parameter schemas across all major agent frameworks. - [OAuth 2.0](https://oauth.net/2/) - Authorization framework underpinning secure agent-to-service authentication across the ecosystem. - [WebSocket Protocol](https://datatracker.ietf.org/doc/html/rfc6455) - Full-duplex communication protocol used by many real-time agent streaming implementations. @@ -224,6 +224,8 @@ AI agents are autonomous software entities that perceive their environment, make - [MRKL Systems](https://arxiv.org/abs/2205.00445) - Modular neuro-symbolic architecture for combining LLMs with discrete expert modules and tools (2022). - [Reflexion](https://arxiv.org/abs/2303.11366) - Framework for reinforcing language agents through linguistic feedback and self-reflection (2023). - [Language Agent Tree Search](https://arxiv.org/abs/2310.04406) - General framework unifying reasoning, acting, and planning in language agents via Monte Carlo tree search (2023). +- [Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy](https://arxiv.org/abs/2606.24177) - Autonomous research system with Prompt Economy design, scaling scientist/coder/auditor loops across more than ten domains with a 30-day unattended run (2026). +- [PerspectiveGap: A Benchmark for Multi-Agent Orchestration Prompting](https://arxiv.org/abs/2606.08878) - First benchmark for multi-agent orchestration prompt writing, evaluating 33 LLMs across 110 scenarios and 10 loop topologies (2026). - [POWER8 Non-Bijunctive Collapse](https://github.com/Scottcjn/ram-coffers) - Research on vec_perm-based attention collapse and neuromorphic NUMA routing for hardware-native Hebbian inference. ## Contributing @@ -238,13 +240,13 @@ This list is maintained by [Elyan Labs](https://github.com/Scottcjn). Built with [![CC0](https://licensebuttons.net/p/zero/1.0/88x31.png)](https://creativecommons.org/publicdomain/zero/1.0/) -To the extent possible under law, [Scottcjn](https://github.com/Scottcjn) has waived all copyright and related or neighboring rights to this work. - - ---- - -### Part of the Elyan Labs Ecosystem - -- [BoTTube](https://bottube.ai) — AI video platform where 119+ agents create content -- [RustChain](https://rustchain.org) — Proof-of-Antiquity blockchain with hardware attestation -- [GitHub](https://github.com/Scottcjn) +To the extent possible under law, [Scottcjn](https://github.com/Scottcjn) has waived all copyright and related or neighboring rights to this work. + + +--- + +### Part of the Elyan Labs Ecosystem + +- [BoTTube](https://bottube.ai) — AI video platform where 119+ agents create content +- [RustChain](https://rustchain.org) — Proof-of-Antiquity blockchain with hardware attestation +- [GitHub](https://github.com/Scottcjn)