Custom agents

Go engine for DAG-based event routing: fan-in/out, backpressure, retry/DLQ, and observability built in.

Prompt context builder for C# projects and AI coding assistants.

Your personal assistant agent for macOS. Your own JARVIS for planning, research, memory, and getting work done.

Ardur — open-source personal AI agent. Lives in your terminal and across your messaging channels. Multi-provider LLM access, tool use, memory, plugins, with cap-token security, receipt-chain transparency, and cost-tuple awareness built in.

Project Friday — personal AI assistant experiment (TypeScript + Firebase, Python backend)

A Project on intelligent study companion for the AI Solvio Hackathon featuring smart notes, PDF analysis, productivity tracking, and collaborative learning with built-in AI tutoring.

Next-gen college platform combining a full-stack student hub (notes, events, placements, forum) with advanced RAG (Retrieval-Augmented Generation) Integration. Features AI-driven syllabus analysis, contextual notice querying, and real-time campus updates.

Universal context optimization for AI coding agents.

Vendor-neutral MIT skills for Claude Code and Codex: cold email, sequences, LinkedIn DMs, humanization, reply-intent, ICP. Approve-before-publish. By Veneris.

A cross-host web-design skill for Claude Code and Codex. Project-derived art direction, a decision router instead of a style pack, and a rendered verification harness that measures the page instead of trusting the code.

Local first AI assistant for your workspace & messaging. Zero data leakage, local storage, and fast CLI controls.

Prevent AI coding agents from acting on stale code context with source-aware freshness checks and hard stale gates.

FastRMCP: The high-level, ergonomic, and production-focused framework for building Model Context Protocol (MCP) servers in Rust. Inspired by FastMCP and FastAPI.

A Vim plugin for inline FIM (fill-in-the-middle) code completions powered by LLMs.

Local-only, agent-agnostic marketing/outreach skill: SQLite CRM, file-approved sends, SMTP.

DSV4 behavior-discrimination benchmark: 81 lightweight zero-dependency pytest grading tests (d10/d11/d12/t2/t3/t4/v4) for datapipe — discriminates harness/agent/persona effects that public tests saturate on

A persistent agent runtime and its official CLI. Three packages: @echo-agent/core (runtime/SDK, zero runtime dependencies), echo-agent (CLI and interactive TUI), and echo-coding (the coding-agent product).

losp - a streaming programming language specifically designed for LLM structured metacognition. It is influenced by Lisp, FORTH, brainfuck, Perl but with a healthy dose of novelty.

Close Code: A terminal-native AI coding agent powered by LLMesh, with multi-model routing, tool execution, context management, and agentic workflows.

Governed cognition substrate for AI engineering agents: mission envelopes, evidence-bound belief/plan graphs, quarantined memory, model-role review, sentinel checks, and BlackFox handoffs under human authority.

No Promises Without Proof. A dependency-free discipline that forces agents to pair every success claim with evidence and a proof state (OBSERVED/INFERRED/UNVERIFIED/BLOCKED), then derives the completion verdict deterministically.

Autonomous terminal coding agent. 15-critic plant, weighted Pareto gate, bounded retry, trunk-based PRs. Bun-native + Zig WASM. MIT.

Xenopus 1.0.0 — a self-improving autonomous AI agent runtime. Local-first, verifiable, model-agnostic.

Self-evolving AI Agent that adapts to user preferences over time. Works with ChatGPT, Claude, and 40+ platforms via Agent Skills open standard.

面向长任务的 Agent Harness 与受控技能自进化框架。基于 Asyncio ReAct 与 stdio MCP，实现任务状态驱动的上下文治理、业务断言 Oracle、有界修补与防退化版本晋升。

Open source agent workforce platform. Next.js + Postgres + MCP server for AI agent teams with human-in-the-loop review, built in.

A tiny but mighty agent that runs in your browser, on your phone, or at the edge.

The best $0 research agent that runs on a laptop. Local Gemma 3 4B + SearXNG + trafilatura. CLI · TUI · Web GUI · MCP server · submittable Claude plugin. 264 tests, all mocked.

A team makes a hard technical decision with AI, and keeps the receipts.

一个将现实议题、规则机制与角色关系转化为可交互多智能体仿真/演绎世界的生成与运行框架。

The first Mac utility built to make your Mac sleep — not stay awake. Sees what's keeping it up, and forces a safe sleep when it should. Free & open source.

A pi agentic extension that stops weak local LLMs from confidently hallucinating. Auto-dispatches a web-browsing librarian sub-agent to verify API defaults, flags, versions & docs from live sources, returning cited evidence. Works with llama.cpp / Ollama. No cloud required.

Give an AI agent full remote control of a Windows PC - with zero console windows flashing on screen. 让 AI 完全操控你的 Windows 电脑，屏幕上不闪任何黑窗口。

梁神按需模式 / LiangShen Lazy — DeepSeek Harness 的按需工具隔离 preset：默认极简，需要什么能力才临时开放什么，用完自动收回。A minimal-by-default, capability-on-demand agent preset for DeepSeek Harness.

Agentic AI code review for GitHub Pull Requests via opencode. Structured, project-aware, remembers previous passes. Works free, no API key required.

Blizz: a nomadic AI development toolshed

A layered, self-hosted Claude Code harness: config + skills, unattended watchdogs, VPS web cockpit (Caddy/oauth2/ttyd/dashboard), and multi-account fleet tooling.

The agentic DevOps tool for you and your agents.

Agentic RAG system using FastAPI, Streamlit, LangGraph, Chroma, Serper, and arXiv search with built-in evaluation and observability.

Transform software development with AI agents that collaborate to deliver high-quality code automatically - from requirements to deployment.

Multi-agent validation for any repository: independent theses, adversarial critics, and a seal whose verdict is computed from real command exit codes — never from an agent's prose. One body, three runtimes: DeepSeek Harness, Claude Code, ZCode.

Developed an AI-powered online exam portal with secure test conduction, automated result generation, anti-cheating monitoring, and performance analytics using React, and Tailwind CSS.

Build Windows apps in 13 seconds from your PHONE. Self-learning local AI agent. Runs on a 4 GB GPU or just CPU + RAM. No API keys, no cloud.

Context at the right moment — not always. No More Agents Dot MD replaces the monolithic AGENTS.md with tiny, security first and declarative context files.

Yunus Pi is a heavily customized harness experience built on top of a forked Pi harness (pi.dev). It uses patches, extensions, skills, special workflows, tools, custom provider configs to enhance the overall experience. It uses advanced machine learning algorithms (wasm/c++), Jev, Cactus Needle 3 and similar technologies.

Agent-deployed, hands-off literature monitor: bioRxiv + journals -> 3-stage LLM filter -> Discord/Telegram digest, with chat-based deep-read & figure follow-ups.

Cross-machine AI coding settings (Mac + Linux)

Compact symbolic notation for Grok-powered agentic systems

100% coverage, still broken. Finds tests that run your code but verify nothing. TypeScript · Jest · Vitest

Local-first, Linear-like task hub for AI agents and agentic harnesses

Never-stopping, always-improving Ralph loop for Google Antigravity — drives the agy headless CLI with a git-backed metric ratchet that guarantees every handoff is strictly better.

Welcome to cerbere-agent guard, originally just agent guard, then rebranded after the mythical hound of the underworld. Our mission? Guarding your AI agents, keeping them in line, and blocking unauthorized actions. Think of it as the ultimate guard dog for AI security and guardrails. 🐕🛡️

Bekon Suite — White list resistant tunnel, GSM voip android gateway, phone control MCP

Make any coding model behave like a mythos-class engineer. A behavioural contract, live in-loop correction, and an independent judge that checks the work against your repo — so "done" means proven, not claimed. Self-validating, evidence-driven agent loops for OpenCode.

go 语言实现的ai 基座+ tui +server系统

Remote MCP server for AfterLaunch. Lets Claude, ChatGPT, Claude Code, Codex and Cursor read AI visibility across ChatGPT, Gemini, Perplexity and Google AI Overviews, read ranked growth moves, ship them and report them live. 58 tools, OAuth or API key.

🐍 让 agent 一直干活到完成的递归 Skill

A Visual Studio extension that integrates OpenCode into Visual Studio.

Telegram/飞书 ↔ tmux AI CLI 双向桥：精确 topic/thread 路由到 Claude Code、Codex、Pi 交互 TUI，支持 WebUI、附件与 systemd

🤖 JARVIS — Fully Local AI Voice Assistant with Iron Man Holographic HUD & Biometric Security (Ollama Phi-4 Mini, Kokoro TTS, faster-whisper)

Attention manager for Claude Code - pauses your media, focuses your IDE, and snaps you back when done

An AI-powered Second Brain Knowledge System that enables intelligent knowledge storage, retrieval, and question answering using Streamlit, FastAPI, and Groq.

Desktop AI workspace for leading models — separate sessions, built-in browser and tools, Studio documents and tables, scheduled tasks, plugins, and MCP. Powered by an open-source core.

🤖 你的 24 小时云端驻留通用自主智能体 (General Autonomous Agent)，由 Google Gemini 3.8 Flash 驱动。全栈开发、情报搜集与自动化中枢，免 API Key，仅 20MB 内存。

Personal multi-model Telegram AI assistant on Cloudflare Workers, with session memory, image input, a chat allowlist and retry handling.

안 물어봐도 미리 해 두는 맥 비서 · A Mac assistant that does things before you ask (app is in Korean)

Iron Man-style AI assistant — voice, memory, GitHub integration, Electron desktop, NVIDIA Nemotron. Built by Nived, age 20.

A bilingual voice assistant for Linux: Whisper transcribes locally, Claude thinks, the shell is its hands

Vibecoded custom windows file explorer

🤖 Python MCP server for AI-powered expense tracking. Built with fastmcp & uv. Chat with Claude to manage finances naturally.

Finexa 💸📊🤖 Smart finance tracking, budgeting, and AI-powered savings insights ✨📈🧠

真正具备工程闭环交付与工业审美的本地多智能体协作与自主开发桌面工作站

CLI tool and MCP server for handling sensitive information without exposing values to AI assistants

Local-first AI operator with persistent memory, governed tools, workflows, voice, and meetings.

Chrome extension that orchestrates multiple AI chatbots (ChatGPT, DeepSeek, Gemini, Grok, Claude, Qwen) — send one question to all, collect answers, aggregate via expert AI. No API keys needed.

A stealthy, high-performance AI Chrome Extension that provides instant code analysis, bug detection, and intelligent explanations via NVIDIA's Llama API. Zero DOM injection. Pure productivity.

Open-source MCP server that runs self-hosted WordPress like a senior admin: 150 tools for content, SEO, security audits, Query Monitor-style profiling, theme.json & accessibility, plugin settings as admin, and multi-site fleets. MCP 2026-07-28, Cloudflare Workers or stdio. Free, no usage caps.

Concept: an operating environment where AI is not an application but the computer's nervous system — organs, afferent perception, a shared scene of state.

Practical CLAUDE.md and AGENTS.md patterns that make AI coding agent rules testable and enforceable.

Claude Code wrote `grep` 4,575 times and `rg` zero times across 25,216 Bash calls in one user’s transcripts. weir is the hook that fixes that — tells the model what shell tools you actually have, suggests rewrites for the antipattern cluster, refuses to run the unambiguously-broken ones (`cat FILE | grep`, `which CMD`).

Editorial AI · 直笔 — A Claude Skill for serious writers. Skeleton-first writing, anti-fabrication, diagnosis+prescription, word count discipline. Built for journalists, analysts, PR teams, B2B writers.

Agentic Coding TUI, not the Ruby gem

AtFlows is a local observability tool for LLM applications. Point your SDK at it, see your costs, tokens, and latency in real-time.

Find the best and newest models for oh-my-openagent (OMO) configurations using Artificial Analysis data

MCP server for live crypto derivatives data from bykaranteli.com: funding, open interest, liquidations, options, ETF flows, Hyperliquid, Solana perps. 68 read-only and account tools; hosted at mcp.bykaranteli.com with OAuth 2.1 or an account key

An AI changed your code. RunBoth runs both versions and tells you what actually behaves differently, including the functions nobody touched.

Event-Driven Cognitive Execution Engine for AI Agents — 612 lines of pure Python, zero dependencies. BM25 retrieval + 5-type classifier + negation detection. Deployed in production for CiteFlow's 3 AI Agents.

Deontic Task Orchestrator for OmegaClaw. Has defeasible event calculus, deontic and temporal reasoning, and agent verifiable work system

Codelicious is a headless autonomous developer CLI that transforms markdown specifications into production-ready Pull Requests with zero human intervention. It orchestrates a dual-engine architecture powered by Claude Code and Hugging Face (DeepSeek for reasoning and Qwen for Coding). Specs -> Code -> Tests -> Iterate -> PRs/MRs

Drive the ChatGPT desktop app over CDP so any terminal agent can consult GPT — your own subscription, strongest model, temporary chats, zero deps. 让终端 agent 随时调用本机 ChatGPT。

open-source skills that make Claude & AI agents genuinely useful — Indian legal docs, GST, startup compliance, pitch decks, cold outreach, content, interviews, data cleaning & more

OpenCode plugin for recall and analysis of past OpenCode sessions: search, grep, and cost/usage insights across your session history via 18 LLM tools over the SQLite database.

Local-first AI resume tailoring engine with deterministic zero-token fact verification, MCP server integration, and sub-30ms Typst WASM PDF compilation. No hallucinations, altered metrics, or broken formatting.

The target-system plugins Covey ships with — Jira, Confluence, GitLab, GitHub, Zammad, Salesforce, Teams, SharePoint, Nextcloud, email, Kubernetes, headless browser — as an ordinary Go module against the public SDK. Shipping with Covey is not a privilege.

A benchmark measuring AI agent reliability on paired Amharic and English tasks in simulated Ethiopian workflows.

Méthode et garde-fous pour travailler avec un agent de code sur un projet qui dure. Extrait de trois projets réels et de leurs incidents.

Natural language → optimized AI context window. Tell it what to build, get exactly the files an LLM needs, trimmed to fit, in seconds.

A local-first, open-source harness for organizing your files with an AI agent, an MCP server for Claude Code, Codex, Cursor. BYO-key, no cloud.

The future of vibecoding.

Typography skills for AI coding agents — 20 focused skills + a /type router, extracted from 10 canonical typography books and reconciled against WCAG 2.2 and current browser behavior.