Custom agents

A high-performance AI coding agent.

Whose Side Is Your Agent On? PrincipalBench — a multi-turn benchmark and post-training methods (prompt scaffold + per-token-KL distillation) for multi-party principal loyalty in LLM agents.

Pre-registered benchmark: HTML dialect + whole-tree rewrite (barkup) vs JSON, granular mutation tools, and two patch dialects, for LLM agents editing typed trees. 9,600 scored runs across 4 models. Findings in REPORT.md.

给 DSH agent 的插件库"起飞前检查单"——任务型请求时自动预检插件库并注入 Top-K 适用插件提示，让插件库利用率可预期、不靠运气。Pre-flight plugin-library check for DSH agents — task-type requests trigger a Top-K hint of suitable plugins injected into the runtime context, making plugin usage predictable instead of opportunistic.

You ask. ANT Acts.

CrewCoder is an extensible, security-conscious coding-agent platform designed for interactive development, multi-worker orchestration, durable sessions, detached goals, and remote execution. Can also be used for the CrewCode ecosystem.

Local-first, LLM-agnostic memory system — hybrid retrieval (vector + lexical + graph), multi-signal ranking. Runs on 4GB RAM, CPU-only, no cloud.

AI-Psychiatry is a Claude plugin that gives AI coding agents an executive-function layer to detect and control attention drift, overthinking, recursive reasoning, compulsive verification, scope drift, retry loops, deadlocks, livelocks, context waste, and failure to stop after completion.

Privacy-first AI assistant for legal documents. Ask questions in natural language, get instant answers with exact sources. Scans contracts for risky clauses automatically. Runs 100% on your computer—no cloud required. Supports local AI (Ollama) or cloud APIs (ChatGPT, Claude, etc.). completely Free

Reliability and safety harness for LLM cloud agents. Tool-selection accuracy vs. tool count, blast-radius guardrails, and prompt-injection red-teaming against a real AWS mock. Every number is a committed JSON artifact.

基于 DeepSeek 的模型自驱多 Agent 深度研究系统:自研 harness、四角色动态编排、GLM 独立裁判评估 + 基线对比、Streamlit 实时编排可视化

Linux kernel dev plugin: L1-L4 verification pyramid (checkpatch/sparse/build/boot) + graph-level governance (Goodhart/blind-spot/conflict) + x-kernel feature audit (P1-P12)

Rust dev plugin: fmt-clippy-build-miri/audit verification pyramid + 12 semantic practice cards + problem-oriented docs map

🐉 HYDRA — 7-headed AI bug bounty system for opencode: hunter, verifier, reporter, plan, auditor, recon, debug · 928 skill modules · one-command setup

🤖 Free open-source coding agent — like Claude Code, but powered by Hugging Face's free Inference API. No paid keys, just pure AI. Built with Python, Gradio, and LLMs.

Sovereign OpenClaw for Android: A high-performance Node.js/glibc engine powering the full OpenClaw experience. Featuring a Flutter UI, real-time 3D VRM avatars, and more

Long-running autonomous hierarchical multi-agent software development engine for terminal, web, and Android. Local-first, multi-provider, open source. Not another a chatty prompt-driven session.

The delivery loop for autonomous software factories

A homebrew Agent framework.

Autonomous Upwork job search agent

An Effect-native agent execution kernel: sessions, runs, turns, steering, follow-ups and typed lifecycle events on top of Effect AI.

Local-first AI Agent framework in Java. ReAct loop, MCP tools, layered memory, multi-agent orchestration — run via Shell, Web or embed it.

OpenClaw Agent Configuration Templates - Lightweight prompt engineering framework for AI agents

A decoupled, zero-dependency dual-track long-term memory engine for LLM AI Agents. | 经验回想与印象整合引擎

Grok Build Remote Web UI ACP Host

Multi-agent HR copilot for employee self-service, policy retrieval, compliance checks, and response synthesis.

A safety layer for AI agents; it decides what your agent is allowed to do - and proves it.

Capri Hub 汇集你的所有 AI Agent

Codex-native Claude review plugin for read-only, evidence-cited Opus review lanes.

Autopilot for spec-driven development: a standalone, opinionated Claude Code framework

Self-host Kinocut + HyperFrames MCP behind Tailscale Funnel, Caddy, and Google OAuth (Docker; no Nix required)

Building an Agentic AutoML system from scratch, step by step, from a simple baseline to an experiment-driven senior ML agent.

Built for AI agents: drive and debug a running app from one config file, and come back with proof a person can watch and an agent can read — not a pass/fail, and not a guess.

Generate a custom newspaper with an AI agent based on your favorite YouTube channels.

Point an agent at a repo + task and run an end-to-end ship pipeline (Go CLI + embedded skill)

An AI engineering team you install into any Claude Code project — stack-agnostic agents plus a deterministic installer. Swap stack packs, not prompts.

Build AI skills that distill any person’s thinking into reusable workflows and prompts

RSI-EAF: XRPL-grounded autonomous factory. LIVE x402 merchant — pay testnet Tag 1/2. https://published-zeta.vercel.app/pay.html

MCP server that brings Perplexity to Claude Code — AI-powered web search, deep research, reasoning, and real-time citations. 9 tools: search, deep research, reasoning, raw search, Q&A, code review, brainstorming, explanations, and file analysis.

Native, local-first coding-agent desktop app in Rust (egui) — run any model: local GGUF, Ollama/LM Studio, SSH-tunneled runtimes, hosted APIs, or Claude Code / Cursor / Codex over ACP. No Electron, no cloud lock-in.

Home Assistant OS add-on for Hermes Agent AI assistant. Includes ingress UI, terminal, dashboard, persistent storage, and gateway support for Telegram, Discord, Slack, and more.

Custom RAG Assistant is a Streamlit app to chat with your own documents using FastEmbed, Qdrant, and Groq LLMs. Ingest TXT/PDF/MD/JSON/CSV, query with context-grounded answers, and switch collection, embedding, and model via .env. Includes handling for timeouts, model changes, and dimension mismatch errors.

Islamic-knowledge AI service for DeenBridge — FastAPI + Gemini assistant with verified Qur'an/Hadith citations, scholar-review fallback, and guidance for Stellar/USDC payments.

DSH Companion 三件套(插件/技能/CLI)发布仓库 - hy-companion 陪伴系统

Stock research dashboard unifying market data and AI-derived news sentiment, with a built-in research assistant for investigating, comparing, and explaining stocks using live sources and cited evidence.

🏠 All-in-one lifestyle management WeChat Mini Program — 17 modules, local-first, dark mode, AI assistant

智密浏览器 Zhimi Browser — 基于 Chromium 源码级定制的国密（TLCP/SM2/SM3/SM4）+ 大模型 AI 浏览器。A GM-crypto (TLCP/SM2/SM3/SM4) + LLM browser built on Chromium.

Production-ready MCP servers for GitHub, Bitbucket, GitLab & Jira - enabling AI assistants to interact with your dev platforms

把 chat.deepseek.com 网页版改造成 dsh（DeepSeek Harness）形态的 Chrome MV3 浏览器扩展 — 复用网页版免费额度，零本地安装

Corrective RAG with LangGraph: evaluates retrieval quality, routes to web search when needed, refines context, and generates grounded answers.

AI-powered e-commerce trust platform with multi-LLM review analysis, fake review detection, and smart product recommendations

MEMO, a visual memory assistant that prioritizes useful uncertainty over confident guesses.

Final project for the IBM SkillsBuild Internship – Agentic AI solution for recommending personalized course pathways based on user interests, skills, and career goals. Includes the final presentation in PDF format.

Stateful local LLM inference manager with persistent KV-cache, native filesystem tools, and instant undo: built in C++ on llama.cpp

A desktop pet that visualizes AI Agent execution states in real-time

Live chat + webrtc video call with AI takeover for multiple websites

Чат с искусственным интеллектом для Android с авторизацией через Telegram/VK, историей чатов, инструментами для работы с изображениями и файлами, экранным помощником, общими чатами и бэкендом на Node.js/PostgreSQL.

支持 VRM/VRMA、语音状态联动与 MCP 控制的跨平台桌面虚拟角色｜中文界面与可配置动作库

个人财务操作系统：账户/流水/负债/应收款管理，Financial Engine 实时计算资产负债，LLM 配置中心 + DeepSeek AI 只读财务分析助手。Vue3 + FastAPI + PostgreSQL + Docker

An AI companion that lives on your desktop and does things — builds native Mac apps on demand, connects to Gmail/Calendar/Slack, browses the web in a real Chrome, controls your Mac, and runs jobs while you are away. Bring any model: Claude, Gemini, DeepSeek, Kimi.

🐾 Pet Care AI Chatbot Application — Interactive virtual veterinary assistant providing pet health tips, diet advice, and symptom assessment.

从 miniVERL 到 verl 0.8.0 的中文交互式实战课程：GRPO/PPO、HybridFlow、ToolAgentLoop 与 VeriCalc

A real Red Alert 2 AI battlefield plugin for DeepSeek Harness with fog-safe control, live DSH Chat, and post-match learning.

Deterministic two-layer validator for LLM tool calls: stdlib structural checker in 25,000-case differential agreement with pinned jsonschema, plus the semantic contract layer where real failures live - qwen2.5-7B: 0 schema violations, 7 semantic, replayed in CI from committed data.

Open measurement infrastructure for agent-facing software. Separate operations from attempts, evidence from inference, and usage from value. Reach → Choice → Use → Utility → Value.

Open reproduction of NVIDIA's AVO paper (arXiv:2603.24517): evolutionary search where an autonomous coding agent IS the variation operator — Vary(P)=Agent(P,K,f). Runs on the Claude Code or Codex session you already have.

Open-source multi-agent governance for running multiple AI coding agents in parallel on one GitHub repo: issue claiming, task market, strict CI/CD gates (GATE-1..6), auto-merge. Install via npx, curl bootstrap, or Codex/Claude marketplaces. Any agent (Codex, Claude Code, Cursor) — no daemon, no vendor lock-in.

适配 DeepSeek Harness 的 OMO 规约智能体与超任务编排工具，用于不间断任务规划、分类委派与自主校验。OMO‑style discipline‑agent + ultrawork orchestration toolkit for DeepSeek Harness, for endless task planning, categorized delegation and self‑verification.

Ask your data. Trust the answer. — A text-to-SQL product for non-technical users.

Interactive viewer and open dataset for autonomous mathematical discoveries produced by Station v2, an open-world multi-agent scientific environment.

Self-improving control kernel for AI coding agents (DAGx AGI Kernel): hard approval stops for irreversible actions, evidence-gated completion, self-learning playbook with drift governance. Behaviorally evaluated, harness-portable Agent Skill.

WordPress performance audit and optimization skills for AI coding agents — Core Web Vitals, caching layers, page builders, and managed-host constraints.

PikoClaw 2026: Rust-Powered AI Agent for High-Performance Devleopment

A curated, always-fresh map of the open-source agentic-AI ecosystem — self-refreshing, ranked by GitHub stars.

Anthropomorphic AI Character Engine: a standardized agent template based on the SPL audit engine for fast, consistent AI character generation. Using causal anchoring and full-logic auditing, it erases OOC, emotional drift and memory loss, offering a stable foundation for AI companions, digital humans, interactive stories, smart customer service and

macOS APFS transparent compression for cold agentic-CLI session JSONLs (Claude Code, Codex, Gemini).

A daemon with a job description. Long-running AI coworkers with sensors, roles, responsibilities and boundaries. No chat window.

Guardrails for AI coding agents — scoped phases, reproducible proofs (exit 0 on the committed tree), and a bundled Multidim analysis engine. Pure Python, zero deps, local-first, Apache-2.0.

AI-powered research assistant & personal project hub

中文 Agent Skill：陪普通用户用 Hermes Agent + Weixin iLink 五步搭建自己的微信 AI 助手，可选知识库、受控 Codex 编程与日常自动化。支持 Codex / Kimi Code / Claude Code / Hermes 安装，默认最小权限。

Chat with your Mac's on-device AI from Chrome — reads and navigates pages for you, runs on local MLX models, nothing leaves your machine

在本地整理 B站观看历史与收藏的 Chrome 和 Edge 扩展。当前为 0.13.0-alpha 预发布。

A personal assistant with chat, long-term memory, daily digests, blog/news monitoring, and paper trading. Built for multi-channel delivery, currently running on Telegram.

Agent skills for building on Vobiz - voice API, SIP trunking, voice XML, and AI voice agents.

Curated awesome list of AI prompt engineering, LLM jailbreaks, prompt injection, system prompt leaks, secure prompting, GPT prompts, red teaming tools & AI security resources.

Agent skill + Python helper for the iKuai router private API — 14 domains, 105 verified func_names

Control Virt-A-Mate from any MCP client: an in-process MCP server shipped as a BepInEx plugin (63 tools)

Native Rust whole-codebase analysis for AI agents.

Capable coding agents on local Ollama models — including the small ones. Task-based model routing, tolerant tool-call parsing, and an auditable permission system.

the agent inbox — a tiny Go CLI for IMAP/SMTP, designed for AI agents

Hosted Instagram MCP server from HasData. Public profile and post-feed tools by handle, structured JSON, for Claude, Cursor and any MCP client.

Yandex Search API plug-in for Hermes Agent

Persistent IPython state, durable snapshots, and native recursive subagents for DeepSeek Harness—packaged as a Prime-compatible Cordis plugin bundle.

Free MCP server to convert, compress & merge PDFs and 100+ file formats (Word, Excel, images, ebooks) from Claude, Cursor, VS Code & any AI agent. Powered by GuruPDF.

Mount a live Pharo image as a local filesystem

Z3-powered reasoning for AI agents. MCP server + CLI for scheduling, logic, config, and safety.

Local-first AI workspace with persistent memory, real tools (files · shell · LSP · MCP · Office), and an opinionated agent team. Your data, your machine.

Hosted Google Maps MCP server from HasData, no Google Cloud project needed. Search places, reviews, contributor history, photos and posts for Claude, Cursor and any MCP client.