Custom agents

SpecFirst é um framework documental em Markdown, distribuído como template, para governar projetos tocados por humanos e agentes de IA.

Autonomous context-engineered development system. 5 agents, 17 commands, SQLite brain. 70-90% fewer tokens. Works with Claude Code, Cursor, Codex, Gemini, Windsurf + 9 more.

A context layer for coding agents: notes, RFCs, tasks and tracked time as one graph of plain markdown. The running app hosts an MCP server, so Claude Code and other agents read what is ready, follow the reasoning behind it, and log their own work.

🚀 Production-readiness skill suite for AI coding agents — 58 commands, 24 security categories, audit/fix/verify across Claude Code, OpenCode, and Codex

Open-core AI bug reporting & testing — right-click to file grounded bugs to Jira/Linear/GitHub/Plane, AI personas (Sims) that review your product, and self-healing end-to-end tests.

OKX Agentic Wallet plugin marketplace for onchain AI agents.

大模型与 AI Agent 实战测试案例集：收录提示词、可运行代码、效果截图与测试记录，按 DeepSeek、ChatGPT、GLM 等分类，支持同题复现与对比。Hands-on LLM & AI agent test cases.

A child repo for your repo — gitignored, shared, auto-synced

⚡ Production-ready Model Context Protocol (MCP) servers, agent skills, and automation workflows for Claude, Cursor, and AgentScope. / 专为现代 AI 智能体打造的开箱即用 MCP 服务与自动化技能套件。

Quota-aware web search for AI agents: 10 providers with local quota tracking (fallback before 429), intent routing, free local-first extraction cascade (gh CLI native for github.com, JSON API passthrough, Russian Trusted CA). Benchmarks included.

Deterministic iOS app validation CLI for AI coding agents.

CLI for reading a local Scrapbox/Cosense-style graph knowledge store

Concordia Simulation Builder provides a web-based UI for creating and running agent-based social simulations using Google DeepMind's Concordia library

Take your AI's memory anywhere - a local-first, portable, git-like memory layer for AI coding agents (MCP-native, $0/offline by default).

Workspace → tab → agent sidebar preset for Herdr, with compact single-tab groups, readable task labels, and provider icons.

Open-source, self-hosted AI workforce. Meetings in, shipped work out. Multi-model AI Council. BYOK.

The **Revenue Recovery AI Agent** continuously monitors business data sources (CRM, billing, support tickets, usage logs, etc.) and proactively alerts relevant teams with concrete, data‑driven recommended actions.

爬虫/自动化 防封指纹伪装（UA+分辨率池）：随机真实浏览器指纹（88+ UA、87 分辨率）降低风控识别。适合爬虫工程师、RPA 开发者、海外社媒多账号。

Auis — code-native design builder: design system, screens, UX flows, review/edit modes, and agent skills for Next.js + shadcn/ui.

A provider-neutral, in-memory agent loop for TypeScript.

Automated production line for self-contained Minecraft adventure delves

A private, native macOS workspace for building and running the ChatBotKit AI platform.

A personal research project exploring whether an autonomous LLM agent can perform L1 SOC alert triage: correlating alerts into campaigns and identifying false positives — without human intervention.

面向二级市场的投研量化与模拟交易 Agent 技能组。支持全流程 SOP 路由、多 Agent 并行投研、量化风控与内置交互式看盘看板。

A multi-loop agent framework for long-horizon research with iterative review, execution-driven repair, and pluggable skills.

Stop Prompting, Start Planning — Reliable Agentic AI with Embabel. Goal-Oriented Action Planning (GOAP) for controllable, explainable, testable enterprise agents in Java & Spring.

Why production AI agents fail without Entity, State, Transition, and Constraint.

Agente de IA que habla el idioma de tu negocio

A foundation for self-evolving AI agents, built on DeepSeek Harness.

TypeScript event history on PostgreSQL for agent workflows and domain applications: typed aggregates, replay, projections, and transactional outbox - zero extra infrastructure

Sergent Specification Golang Reference implementation

ZKP-RA: Zero-Knowledge Proof Reasoning Anchors for mitigating memory poisoning in autonomous agentic DeFi — Groth16 zk-SNARKs, Circom circuits, on-chain Solidity verifier, Python agent guardian. 182ms proofs, 9.7x gas reduction vs zkML.

E-commerce AI Agent - Agentic RAG, HITL, Skills

Sergent Specification TypeScript reference implementation

Sergent Specification Python Reference Implementation.

Open-source AI penetration testing tool to find and fix your app’s vulnerabilities.

Example apps built with Sergent Python reference implementation

SynROV is a multi-robot platform. The project combines Arduino firmware, a Processing-based 3D operator station, a browser control console, the Python AiBot runtime, and optional ROS 2/rosbridge integration.

Pixel Perfect for harnessing image comparison. The result looks close, but the spacing is off and now you need to explain what to fix in plain english to your agent, good luck.

Open-source business engine for AI agents. Invoicing, payments, inventory, checks and cash as permission-controlled MCP tools — with idempotency, structured errors and an immutable audit trail. Self-hosted, offline-first, multi-country.

Sergent Specification Rust Reference Implementation.

No GPU, no LLM — agentic RL from scratch in pure Python. 不用 GPU、不用 LLM，纯 Python 手搓明白 Agentic RL 训练闭环（trajectory / reward / loss mask / GRPO）

Point your AI coding agent at this repo and it installs a personalized ruleset for reproducible, analysis-first data science in R and Python — environmental, ecological, and geospatial.

Open-source, AI-operated platform that makes deploying and running web apps on your own server as simple as using a managed cloud.

AI-assisted Aspen HYSYS skill for existing-case takeover, validation, and reporting. AI 辅助接管已有 HYSYS case，完成验证、导出和报告。

AI Agent 驱动的 Primavera P6 施工进度计划生成工具 | CSV → XER → P6

元构·智能生命体（MetaGO Intelligent Lifeform）— 守律 · 进化 · 可溯的智能体运行时。112 技能 · 162 MCP 工具（6 服务器/核心 112=55+57）· 927 算法 · 38 公理 · 47 属性 · 121 协议 · 161 引擎 · 597 原子 · 11 平台 · Engine 2.1.3

Extract complete, reviewable figures and tables from PDF, arXiv, and HTML papers — an installable AI-agent skill and CLI with contact-sheet QA.

261 Kimi Agent core skills covering writing, research, finance, product planning, engineering, data, documents, media, and productivity.

中文自然语言问数的电商数据分析 Agent：LangGraph 编排 NL2SQL 链路 + sqlglot 安全校验 + RAG 知识库与指标语义层 + 双轨评测（开发集 EX 95%）+ XGBoost/SHAP 归因 + 流失/履约诊断工作流 + MCP 标准化工具

AI Agent Virtual Office — pixel-art workspace for watching agents work in real-time via MCP

Governed skills for AI agents: reusable workflows, risk checks, and deterministic validators for Codex, Claude, Gemini, and more.

专为AI Agent设计的轻量级思维树+代码深度分析插件。两级触发链（开启思维树→开启深度思考模式），耦合基因库省Token，支持5语言函数级分析、安全/性能/逻辑检测、沙箱动态测试。零依赖，手机端Termux流畅运行。

One MCP server for every Shopify store. Search products, compare inventory, find catalog gaps, track fulfillment SLAs, and make guarded updates from Claude, ChatGPT, Codex, Cursor, and VS Code. Host it for your team with per-user Shopify sign-in.

《AI超级个体》一人公司（OPC）创富手册配套 Agent Skills

AI-native issue tracker with MCP integration, human/agent time tracking, and velocity analytics. A self-hosted alternative to Jira and Linear for developers who work alongside AI agents.

Claude Code skill: design multi-agent workflows as dependency graphs, not linear pipelines

Open-source AI recruiting platform for resume screening, evidence-based candidate evaluation, interview planning, and AI interview orchestration.

R-CAGE: Recovery-conditioned agent governance for zero-trust, ransomware recovery, software supply-chain security, and crypto-agile agentic defense.

The Open-Source Astra 6 for your PC. Your AI can see, click, type, navigate, and get real work done on Windows.

Forge Studio - Agentic SWE factory builder

Melo: The AI-Native Voice Agent. Your Intimate Voice Buddy!

Production code for the book "LLM Inference Engineering: Latency, Cost and Reliability at Scale"

Agentic multi-agent framework for autonomous LLM vulnerability discovery: CVSSv4.0 scoring, 8-framework compliance mapping, SARIF output. OWASP LLM Top 10 (2025).

Agentic QUantum Interface for Load Allocation: backend-agnostic quantum-classical interface for optimization offload in agentic AI, with controlled QAOA evaluation and a post-quantum-secured (ML-KEM-768/ML-DSA-65) agent messaging layer.

TRACE-MAS: Unified cryptographic framework securing multi-agent LLM pipelines against drift, spoofing, prompt injection, and temporal attacks

Lean coding harness written in Rust, focused on fast local workflows, deterministic tooling, and efficient code retrieval.

Agent-native bi-temporal graph store: verified temporal operators as LLM tools + trace-grounded answer verification

seanswarm: multi-agent workflow pack for coding agents (Claude Code / Codex / Cursor) - evidence-governed research fan-out, dual-read docs, dev delivery with independent review, and visual acceptance gates.

Nexus — AI Coding Agent Desktop App. Free, local, open source. Actually builds things — not just chat about them.

Build with AI. Run with guarantees. TPF is a Java framework for strongly typed business flows that compose model decisions, authorised SaaS capabilities, and deterministic code—with durable effects, waits, retries, replay, observability, and flexible deployment.

Context-aware AI desktop assistant for macOS

🤖 MahoAI: Full-stack AI coding assistant and automated backup/refactoring environment with Flask & modern web interface.

An end-to-end voice cloning processing pipeline featuring an asynchronous job queue (Celery/Redis) to handle voice cloning , referring, import through readymade api"

个人 AI 工作台，让 AI 自然协作于你的电脑：对话、文件、终端、浏览器与可扩展应用。A personal AI desktop with chat, files, terminal, browser, apps and widgets.

触手怪 BrainTentacle · 桌面 AI Agent 中枢 — 对话 / 工具 / 记忆 / 技能 / 安全脱敏 / MCP 扩展，Tauri + Rust + React，覆盖 Windows·macOS·Linux·龙芯·RISC-V

Scanned-document workbench that AI agents can operate: clean, OCR, annotate PDF and DjVu, with a local MCP server. macOS, Windows, Linux, Web.

這是Twitch的聊天室機器人，透過LLM模型，使用主播的個性代替主播回覆留言

Python CLI for Facebook & Messenger automation — post, reply, send messages, manage comments via GraphQL API + Playwright browser automation with E2E encryption support

A Personalized AI that tracks the user behaviour over problems and tries to monitor the knowledge of the user.

Open-source agentic platform to build, deploy, and scale agents that vibe with people.

A JavaScript web app delivering daily Coptic Christian content: Agpeya prayers, saints, Bible verses, and Desert Fathers sayings. Features single-click buttons with localStorage persistence, Coptic date and liturgical tracking, JSON data, and an optional AI chat iframe.

This repository contains the implementation of a chatbot that helps in answering user's questions regarding the restaurant CapreDiem and helps users for online restaurant reservation.

🤖 Analyze chatbot conversations to enhance performance and user satisfaction through data analytics, clustering, and predictive modeling.

Typescript/React Library for AI Chat💬🚀

Compendio is an MCP server that indexes a project's markdown documentation and exposes it to any AI agent.

制造业 RAG 与设备运维智能体：支持企业知识问答、故障诊断、工单生成及 FastAPI 系统集成。

Secure, local-first document-to-Markdown conversion. Zero-trust input sanitization, multi-column PDF layout, LaTeX math, PII masking. No cloud dependencies.

A predictive AI learning platform helping students and institutions prepare smarter through intelligent study sessions, exam pattern analysis, and trusted educational resources.

Fully local MCP retrieval tool for coding agents. 161M encoder, ~1s per tool call, no API, no LLM in the query path. No file reads: line numbers, callers, callees and exact call sites, packed to a token budget. 2-4x the file coverage of BM25 and TF-IDF on SWE-Explore, and tops the paper's own sub-1B code model on CORE-Bench.

Local-first persistent memory plugin for opencode: three-layer memory (long-term knowledge / episodic summary / working memory) with SQLite + Chinese-aware retrieval (RAG), fully local.

A private AI companion that reads your journal, remembers you, and finds your behavioural patterns — 100% local with Ollama. Nothing leaves your machine.

OjuIye is a provenance-verified, multilingual fact-checking system for Nigeria's 2027 elections. Uses cryptographic source signing to defend against RAG corpus poisoning, with hybrid retrieval across English, Hausa, Yoruba, and Igbo.

🍳 Your AI second brain. Self-hostable. Get answers from the web or your docs. Build custom agents, schedule automations, do deep research. Turn any online or local LLM into your personal, autonomous AI (gpt, claude, gemini, llama, qwen, mistral). Get started - free

Portable hybrid (BM25 + dense + RRF) retrieval engine and a label-free evaluation harness — extracted from a personal AI-assistant memory index and decoupled to run on any source tree.

Search your Mac by what a page looks like. ColQwen2 over page images, everything local.

面向计算机考研 408 的本地优先智能学习助手：以 RAG 全链路可溯源问答，内置四科知识点与 1175 道选择题，AI 出题即时判题；自带 200 题量化评测体系，Docker 一键部署，学习数据全程本地运行。

Local memory + semantic search for AI agents. Indexes code and documents on your machine, queryable via MCP. Private, offline, fast. macOS arm64 + Linux x86_64.

Self-hosted AI audio analysis & knowledge workspace for transcription, speaker/emotion review, business insights, and RAG.