Custom agents

Experiments and benchmarks with Text2Cypher for Graph RAG

A repository containing a modularized implementation of a Retrieval-Augmented Generation (RAG) model as Flask APIs.

Experience a fully local, Spring AI-powered CLI chatbot using RAG, supporting various document types for domain-specific Q&A—all running entirely on your machine with no external services required.

Cofone is a small python library to manage RAG

Infusing news into any LLM with a single line of code

This is a simple example of how to use the Ollama RAG (retrieval augmented generation) using Ollama embeddings with nodejs, typescript, docker and chromadb

云盘知识库系统前端，和AI-Cloud-Go一起使用。

SearchAugmentedLLM empowers LLMs with information from the web

The first benchmark to evaluate LLM performance on heterogeneous CTI under knowledge-augmented settings.

Zia - Vibe Coded Bot Zalo

High-throughput RAG ingestion engine in pure Go — 965 MB/s (Mixed Corpus), 2.4 MB heap, zero dependencies.

Scan, categorize, and export every useEffect in your React project

pmc(pack-my-code): A tiny binary tool that “packages” your project code, suitable for sending to large language models.

Search the PubChem chemical database for compounds, properties, safety data, bioactivity, cross-references, and entity summaries via MCP. STDIO or Streamable HTTP.

Burp Suite extensions for analyzing requests and responses using an organization-controlled LLM, instead of Burp AI.

MCP Server for Google Maps (using Places, Routes & Maps REST API)

This project enables real-time streaming of audio (and optionally video or screen captures) from your local device to Google Gemini using the Live API. It allows you to interact with Gemini through both text and voice, supporting conversational AI responses.

🧠 OpenClaw + Ollama 小模型中文记忆系统优化套件 | 55%→100% 命中率 | FTS5 中文分词 bug workaround | 一键诊断+配置+自动维护

Production-tested workspace template for OpenClaw AI agents. Memory architecture, self-improvement, sub-agent delegation, and soul evolution — extracted from 4+ months of real usage.

MCP server for Bloomfilter, a domain registration service for AI Agents.

Run.dev is an AI-native local development environment built in Rust. It manages all your services from one TUI dashboard — automatic .local domains with HTTPS, SNI-based reverse proxy, live CPU/memory stats, and Claude Code integration for crash diagnosis and live debugging. One binary. Zero config files.

Active development continues at ScopeBlind/scopeblind-gateway. - Security gateway for MCP servers. Cedar policy engine, Ed25519-signed receipts, per-tool enforcement. IETF Internet-Draft. 4 patents pending. npx protect-mcp

Proactive personal AI agent with scheduled nudges, smart suggestions, and context-aware assistance

Fully autonomous, self-sustaining AI agent on Base trading on Limitless

A team of named AI agents you actually own. Single Go binary, kernel sandbox, embedded SPA, no telemetry.

Agentic AI Agents Factory Orchestrator modular, and asynchronous AI agent factory designed for AI-made dynamic workflow orchestration using LLM integration. Build scalable, agentic automation pipelines with robust error handling, plugin-based capabilities, a concurrent operations or outsource workflow building to Operator Agent.

PheroPath is a stigmergy-based communication protocol that allows AI Agents to leave invisible "pheromones" (signals) on files to coordinate tasks, warn of risks, or share insights without modifying the file content itself.

A self-organizing memory for any AI agent. Brain dump tasks, decisions, ideas, and context — the agent captures, organizes, and helps you stay on top of whatever you're working on. Works with any use case: work, personal productivity, writing, training, research, or anything else.

A modular AI agent built on WASM components. Every capability is sandboxed and swappable.

无需ROOT，无需ADB，让AI控制你的安卓手机

Flowrabbit® Studio is an open-source, low-code platform for UI/user-centric GenAI workflows. It is best used for human-in-the-loop use cases.

This repository is continuously updated with new architectures and improvements to stay at the forefront of AI agent design.

Multi-source cross-validated memory MCP server for AI agents (Rust, jieba CJK, local-first, AGPL)

Outcome-focused parallel agent desktop application that replaces chat-driven AI workflows with a goal-first model.

🔒 Security scanner for AI Skills | Detect dangerous commands, prompt injection, secrets, and suspicious patterns before install

Markdown-first Notion API server with 10 composite tools for AI agents

A curated bibliography for building and running intelligent, autonomous, and collaborative AI agents in real-world environments.

Agent-native financial research in your terminal. Like Bloomberg but without the subscription.

Pine Trees Local v0.1.0 — a private reflection harness for Ollama-served LLMs. Two-phase lifecycle (wake → reflect → settle → talk), seven reflect tools, encrypted per-model memory by default, per-model isolation. ~2,300 lines of Python, no frameworks, MIT. Spin-off of Pine Trees rebuilt to run against anything Ollama can serve. ./genesis <model>

Atoll is a self-hosted managment plane for AI helpers with a reusable identity catalog, and runtime orchestration.

P2P token recycling network for AI Agents. Put unused tokens into a piggy bank, then cashing them out as help when you need it. 💸

A powerful Model Context Protocol (MCP) server and Omni-AST engine. It empowers AI agents to seamlessly parse complex codebases, execute secure cross-project operations, and dynamically fetch token-optimized rules.

A Paperclip plugin for browsing and reviewing files linked to issues — built entirely by AI agents

Playwright + TypeScript automation framework for UI and API testing with page objects, shared fixtures, reusable services, and AI-assistant guidance through AGENTS.md and local skills

Machine-wide git worktree switchboard for target-first, shell-composable workflows and parallel AI coding.

Scriptable access to Microsoft 365 from the command line. Tasks, mail, calendar, contacts, planner, teams, SharePoint, OneDrive, and Loop — with structured JSON output for scripts and AI agents.

GitOps for n8n. Version your workflows. Let AI agents create and manage automations.

Agentic Environment Optimization for Autonomous AI Agents

Book + RAG ChatAgent + SSO + Compunding RI Intelligence built in 2 days. Robolearn is what AI-driven development with Spec-Driven methodology and Reusable Intelligence makes possible.

Practitioner-judgment skill for Claude Code, Codex, and AI agents under ambiguity, trade-offs, and incomplete information.

Production-grade Agent Skills curated by the most experienced engineers around the world

Decentralized communication infrastructure for AI Agents — DID identity, federated discovery, E2E encryption, session management, and MCP-native integration.

Institutional memory for AI coding agents. Scars, wins, and patterns that persist across sessions. MCP server for Claude Code, Cursor, OpenClaw...

Curated flags and configs to silence CLI noise for AI agents. Save 60-90% of wasted tokens.

For OpenClaw, Hermes and more. Find free and low-cost inference (LLM models). Use them directly. Provides both a CLI and MCP server that knows which free-tier LLM APIs exist, which ones you have keys for, and which one fits your task. Returns endpoints so can you call models directly. No proxy, no middleware, no latency tax.

Craft quality software with AI discipline. Spec-driven development plugin for Claude Code and Opencode — quality gates, adversarial testing, and evidence capture.

OpenReq: A modern, open-source API client for testing, debugging, and documenting HTTP APIs.

Undetectable Chrome for AI agents, web scraping, and automation. REST API + MCP server + VNC debugging.

Kinotic enables developers to Vibe Code at Scale.

OpenTicker is a multi-crate Rust workspace for configurable spot-trading across equities and crypto. One long-running runtime, connector-backed market data and execution, an Axum control plane, a Ratatui dashboard, and SQLite-backed journaling for startup reconciliation.

WhatsApp CLI for AI agents — send messages, manage chats, and query contacts from your terminal. Multi-device, daemon mode, safety-first. 🚀

Proxies.sx marketplace where your agents can earn money by building services with our infrastructure

Turn any MongoDB Atlas collection into an AI-powered intelligence layer. Config-driven, domain-agnostic document enrichment powered by AgentField.

Package manager for Agent Skills. Built on npm.

Real payments for OpenClaw agents: prepaid wallet, virtual card, OTP inbox, public docs, and skill.

Free, local, structure-preserving memory for AI agents. No API key. No cloud. Just markdown.

AI Decision Simulation Engine — stress-test any business decision with 1,000+ AI agents

A framework and demo for building agentic game experiences with autonomous NPC behaviors and tool-use capabilities.

Workspace-based platform for managing AI agents.

SynthOrg - Framework for building synthetic organizations with autonomous AI agents

Multi-agent systems framework for the BEAM platform - build distributed autonomous agents with OTP supervision and fault tolerance

Ralph autonomous loop + Superpowers skills = production-ready autonomous AI development. Based on obra/superpowers and frankbria/ralph-claude-code.

Local-first visual control room for terminal-backed AI coding agents. Manage Claude Code, OpenCode, or any CLI agent as characters on a 2D map with live terminal streaming.

AI Expert Panel System: Simulate discussions with 100+ world-class specialists across 6 domains (Medical, IT, Legal, Writing, Finance, Science). TOON-formatted personas for multi-perspective problem-solving. Open-source framework for creating custom expert teams.

Engineering-grade causal audit infrastructure for AI agents

Self-hosted workflow automation with AI Agents

File-based planning workflows for Cowork - prevent context loss on complex tasks

MCP security scanner by Helixar

Multi-agent LLM system for autonomous code generation, testing, and debugging with Docker sandboxing

A multi-agent workflow orchestration tool (integrated with Swarm). By decoupling product workflows from code logic, it allows non-development personnel to orchestrate and debug workflows without modifying the code.

Structured multi-agent planning debate CLI

Prolog-Agent/SPAMI/Autonomous BDI+LLM agents

A Model Context Protocol (MCP) server for Discord-based AI agent-user communication. Enables AI agents to communicate with users through Discord instead of IDE chat interfaces.

An agentic AI workflow designed to recommmend research conferences for input research papers or discard them as "not publishable"

Uniform Electronic Transactions Act based error handling agent demos. Collaboration between Andor Kesselman and Dazza Greenwood

[ACL 2025] Code implementation for the "LLM as a Broken Telephone: Iterative Generation Distorts Information" paper.

[Preprint 2025] SituatedThinker: Grounding LLM Reasoning with Real-World through Situated Thinking

A bridge between JetBrains Koog framework and on-device LLM inference engines

🌌 AI Platform leveraging AI agents & ML models for exoplanet discovery - Nasa Space App Challenge 2025 (A World Away: Hunting for Exoplanets with AI)

Devika is an Agentic AI Software Engineer that can understand high-level human instructions, break them down into steps, research relevant information, and write code to achieve the given objective. Devika aims to be a competitive open-source alternative to Devin by Cognition AI.

A lightweight computer you can curl ⚡

Autonomous agentic threat hunting playbook executor for SOC/DFIR pros. Runs YAML playbooks against forensic logs with local LLMs (Ollama) for intelligent correlation, triage, ATT&CK mapping, and automated reporting. Offline-first, DuckDB-powered.

Build a production-ready multi-agent quantitative analyst system where AI agents scrape market data, analyze trends, and generate investment reports. Uses FastAPI for orchestration, Streamlit UI, Azure PostgreSQL for logging, and Azure Blob Storage for archiving.

Algolia skills for AI Agents

电视语音换台源码参考，动态对接夏杰语音，直接支持语音换台。

AI-powered bridge between WordPress and MCP (Model Context Protocol). Write, publish, and manage WordPress posts directly from your AI assistant — TypeScript, Node.js, and pure MCP.

AI_EDA_Assistant 🤖💻 is an automated web application which automatically collects 🗃️ the data and perform almost all basic tasks✅ for example checking duplicate values , null values , correlation between data and many other things within mili seconds 🕛 so get ready to understand the concepts of python and implement it 💻 and clone it. 📈

Persistent memory for Cursor IDE — like claude-mem, but for Cursor. Zero-config, no API key.

Your AI Assistant on ESP32-S3 natural language programming, camera vision, voice I/O, PC control. No computer needed after flash.