Custom agents

Hermes-style persistent memory (MEMORY.md / USER.md) for DeepSeek Harness (DSH) — a faithful port of hermes-agent MemoryStore: bounded, file-backed, model-curated, frozen-snapshot injection.

An issue tracker you run yourself, built for teams and the AI agents they work with.

Self-hosted, AI-agent-native governed data platform with ontology, policy controls, lineage, and auditable writebacks.

A contract-driven multi-agent pipeline for researching, designing, and building client websites with AI.

Windows-local MCP server and Dalamud plugin for observing FFXIV state and executing opt-in actions.

Command-line tool for finding, inspecting, citing, and accessing 350+ ophthalmology datasets from official sources.

CLI toolbox for Observability-Driven Development (ODD): coding agents observe local runs on an OpenTelemetry/Grafana stack - or remote ones on any OpenTelemetry backend - and feed the next spec-driven improvement loop.

43 Claude Code skills for forward deployed engineers: land in an unfamiliar codebase and carry work from requirement to go-live, on any stack. Cold-start tested against 3 stacks.

Feature-rich WhatsApp bot powered by @whiskeysockets/baileys and lowdb. Packed with interactive mini-games, gacha waifu system with trading mechanics, media tools, and automated group management.

Open source AI astrology chatbot. Vedic kundli, Western charts, BaZi, feng shui, tarot, numerology, dreams. Real calculations via remote MCP, multi-LLM. Next.js 16, MIT.

An agent-skill & plugin to render programmable visuals (images & animations) for Claude Code, Cursor, Copilot, Codex & others

Role-aware, permission-enforced benchmark for LLM agents that operate HPC systems — SLURM, telemetry, RBAC, docs, facility. A policy violation hard-fails the task. 88 tasks (10 categories x 5 roles), 29 reproducible environments with 6 rebuilt from real Marconi100 ExaData, scored on 7 weighted dimensions.

Governance-as-code for AI coding agents: author a policy once, enforce it on every agent (Claude, Copilot, Cursor, Codex) via git hooks, CI, and native controls.

The Raspberry Pi desk dashboard you can talk to: widgets, touch, and a voice button wired to any AI including one running entirely on the Pi. No accounts, no cloud.

把 Claude Code、Codex 和 Hermes 接入工单、CI 与业务系统。自托管 Agent 执行服务，提供 Web 管理台、异步任务 API、独立工作区和结果回调。Self-hosted agent execution with a web console, async task API, isolated workspaces and signed webhooks.

AI compliance scanner for Meta, Google, TikTok, and Taboola.

Tael is a payment layer that enables autonomous AI agents to securely pay. Won 21 days stellar buildathon with Tael.

Turn your codebase and GitHub PRs into a live isometric city where AI agents visibly build and review your software in real time.

One sentence in. A polished, editable PPTX out. We reverse-engineered frontend patterns behind Claude Design and Fable for visual design and human-intent QA.

Local-first MCP server: give ChatGPT & friends grounded, current context on the repos your coding agents work in. Live git, docs, tasks, agent handoffs.

♾️ Doubao Work alternative, chat agent to web apps with various UI and safe, quick, stable APIs. ♾️ 豆包工作开源平替，聊天即兴生成各种网页前后端，安全、快速、稳定。

Safe, local-first internet access for AI agents — zero login, zero cookies, no third-party data services. Install once to read public web pages, YouTube, GitHub, Bilibili, V2EX and RSS.

LLM-optimized Schema Catalog for the Kubernetes Ecosystem

Keep your Mac awake with the lid closed — without burning battery. A menubar app for AI-agent and long-running-build sessions.

Developer skills for Univer aimed at AI agents, providing accurate guidance for integration, extension, and development.

Local-first, evidence-led security knowledge base with a model-neutral JSON CLI, immutable sources, and human-reviewed proposals.

Agent-ready creator strategy kit that turns brand truth and breakout content patterns into creator archetypes, hooks, concept slates, and test plans before sourcing or production.

An autopoiesic kefir for computing machines

Monolingual Italian late-interaction (ColBERT) retriever for search and RAG

First-Person Agent Memory Bench. 10 Categories including fact recall, multi-hop links, temporal reasoning, fact overwrites, speaker traps, refusal, credibility, and agentic tool usage. 540K token / 60 session corpus, all in first person. Dynamic output-answer-key portion. Comprehensive report with visuals and miss breakdown.

Zillow MCP server by HasData: for-sale, for-rent and sold listings and full property details, as JSON.

Ziwei Doushu (Purple Star Astrology) — Claude Code skill for Chinese destiny analysis. Multi-school: San He, Fei Xing, He Luo, Qin Tian Si Hua. Discover ancient Chinese metaphysics, Taoist astrology, Eastern fortune-telling. Powered by @mingai/mcp. 紫微斗数 算命 命理

AI 村口情报中心：AI 情报日报 + 截止日期救命器，由 AI 与飞书共同完成持续运行的个人情报闭环。

Turn any URL into structured intelligence — SEO, tech stack, contacts, products, social profiles, security headers & AI-ready Markdown via one fast REST API.

Become an AI Engineer in 24 weeks - hands-on, project-based roadmap for software professionals. No theory hell. Build real AI apps from week 1.

Multi-agent harness for CLI coding agents — orchestrate Claude Code, Codex, and more.

Full local control of your Zotero library — from your terminal or your AI agent. No cloud, no API key. CLI + MCP server + bridge plugin.

A digital twin that learns what you'd want — and does it. Delegated judgment with safety constraints, explanations, and progressive trust.

AI agent for video production — text/image-to-video generation, avatar/UGC talking-head videos, and data-driven ad-creative remakes, powered by muapi.ai.

Open-source CLI that recovers a web app’s shipped source, rebuilds its tree, and exposes what crawlers miss. For authorized security research.

BlackPearl — DeepSeek-powered AI agent harness and orchestration ecosystem with 286+ skills, 273+ specialist agents, plugins, security, QA and automation.

Structure-aware Markdown chunking for RAG over visually rich documents: keeps tables, headers, fences and figure captions intact, with a reproducible structural-fidelity benchmark.

Personal cognitive-retrospection Agent for tracing viewpoint changes across long-term Obsidian notes with source-grounded evidence.

A pixel-art office where AI agents actually do the work.

Multivariate ad creative fatigue auditor & refresh triggers (AdEspresso style)

MCP server for semantic hybrid search (sqlite-vec + FTS5 with optional cross-encoder reranking) over a Markdown / plain-text knowledge base

LLM specialist in AWS, GCP and Azure acting as a senior cloud engineer — RAG + read-only tools + domain specialists

AGENTIC · OPERATIONS — A second brain for your company, and the team of agents that runs it. Duplicable Company OS template: one question, a bus, two registries, tiered autonomy, self-healing loops. Zero shared keys.

Memory Layer for AI Agents

Nerve Center - OpenClaw Work Center

Local-first desktop control room for coding agents. Run Codex, Claude Code, Cursor, Grok, Mistral, Pi, and local models together — each with its own role, permission posture, and receipts. Multi-agent ensembles in one thread, diff review, on-device history. macOS, Windows, Linux.

Agent wallet infrastructure

Production-grade Blender extension and cross-platform asset pipeline for validating, optimizing and exporting mobile-ready 3D assets to Godot, Unity, Unreal Engine, Flutter and Android.

Claude Code skill that explains code changes in plain human language — no scary jargon. Two modes: a 'for dummies' story or a per-file walkthrough.

tmux-style multiplexing for Neovim: every project is a detachable server with warm LSPs and live terminals, plus an agent drawer.

DeepSeek Harness 白皮书 · 131 页入门加进阶综合教程：从装上它、看懂它，到自己扩展它。面向想深入研究 agent harness 的开发者，所有结论回源码核对。

Context Engineering完整中文教程：RAG系统、记忆架构、多智能体、认知工具、神经场论 | 750K+字 | 即用模板

SECS (SECurity aSsistant) turns an AI coding agent (Claude Code, Cursor, Codex) into an authorized information-security assistant - 35 security Agent Skills, a Debian/Ubuntu/Kali toolchain installer, and a local Metasploitable 3 practice lab, under strict rules of engagement.

Beginner-friendly Sol-led orchestration for Codex with pinned Luna and Terra workers and verified review.

GKG is Pre-indexed code knowledge graph for Claude Code, Codex, Cursor, OpenCode, and Hermes Agent — to save 💰💲, 🌟 Star if you like it! 🌟

Plot anything with zero effort. Editable charts, diagrams, figures, posters, and slides.

Use Google Antigravity CLI as a guarded Codex subagent through MCP

Your coding agent says it works. Gopnik tries to prove it doesn't — against the code, the delivered revision, and the checks themselves.

LLM 驱动的求职 AI Agent：8 平台搜岗 + 精排匹配 + 简历定制 + 材料审核 + 投递追踪 + 模拟面试 + Offer 评估 + 薪资谈判，本地部署

An agent skill for coding agents (e.g., Codex, Claude Code) that conducts structured UX research discovery interviews, guiding users from vague product ideas to clear discovery insights.

24 free SKILL.md skills for Claude Code & compatible agents: qualify leads, research companies, clean lists.

AI-first parametric CAD and 3D-print preparation from chat, voice, parameters and images.

World's first compiler and execution platform for AI systems — turn prompts, skills, and multi-step agent workflows into versioned, packageable software that runs with dependency-aware parallel execution.

The fleet and facilities console for agent runtimes: attach to runtimes you already run, or provision new ones, across machines and harnesses.

Open Source Instinct — a personal AI agent you text to get things done.

AI 远程诊断客户电脑与 Android｜Self-hosted AI remote diagnostics via MCP and CLI.

Converge — Dual-agent iterative convergence for plan/code review | 双 Agent 迭代收敛器，用于计划与代码的对抗式审查

MCP server for UnknownCheats and Elitepvpers - search, threads, Cloudflare bypass

Give your coding agent a workflow for chip front-end design and verification.

All the various messaging channels for OpenHuman

A portable, self-hosted memory plane for AI agents — one core across API, MCP, CLI, and UI.

awesome-agent-engineering(本地目录 ~/AAE)

Your backlog, shipped in parallel — a desktop app that runs Claude agents across your repo's tickets, each in its own git worktree. Triage in, PRs out.

Geocode places, fetch global weather forecasts, historical climate, marine conditions, air quality, and terrain elevation via MCP. STDIO or Streamable HTTP.

Local-first AI brain for personal relationships and communication, powered by MCP｜多平台身份图谱、关系记忆、回复策略与受控自动回复

Evidence-first legal workflow for Russian civil practice, with intake, drafting, review gates, document assembly, and privacy controls.

Personalized video sales outreach script & storyboard planner (Tavus style)

Your trusted advisor for managing knowledge, projects, and ideas — powered by AI

CLI coding agents for everyone — checkpoints, preview, plugin/skill/MCP management, AI teams.

Non-LLM AI substrate that remembers who it is across sleep gaps. Pure numpy, Hebbian + sleep replay + persistent slow weights. Ten identity tests, two audit rounds, contributions welcome.

Spine — governed, provenance-grounded autonomous software delivery (installs as agent-orchestrator).

Скажи словами, что построить — получи работающую версию, проверенную до того, как её назовут готовой. Плагин для Claude Code: сам собирает требования, план, код и приёмку; «готово» закрывает код возврата, а не фраза в чате. Откат одной фразой, стоп одним словом, без терминала.

Rust native Alerting and Incident Management on your CLI & favorite Coding Agents

A browser-based OS where every app's UI is hallucinated live by AI. Inspired by the Microsoft Build 2026 Vibe OS demo.

An instruction library for agentic workflows

Kaval.AI is an opinionated Python library for building well-defined, testable and robust agentic workflows, chatbots and tools.

MAC版灵动岛--Agent运行监测：本地查看 Codex、Claude Code、VS Code 等 AI Agent 的运行状态、工作时长、对话与 Token 用量。

Five layers of an agent harness, each with the failure it prevents. Runnable with no API key.

Enterprise-grade AI Decision Intelligence platform. Features an automated AutoML engine (XGBoost, LightGBM, CatBoost, LSTM), SHAP explainability, Gemini-powered AI Copilot, FastAPI microservices, and React/TypeScript frontend with end-to-end CI/CD and Docker orchestration.

Sample projects repository for Syn Bot Channels.

Audit every skill installed on your agent. Finds the ones that can never fire, measures what they cost you on every message, and flags any shipping runnable code.

40 curated agent skills — plain SKILL.md files for building calm, native software and running AI coding agents well. Claude Code, Codex, Cursor, Grok, Goose.

Orbit is durable, intent-tracked agentic project management for developers

Most AI will tell you your startup idea is brilliant. This one researches it properly and scores it out of 100. The two ideas I have put through it came back 43 and 28

Machine-enforced frontend UI/UX skill pack for AI agents. Registry + lazy loading: a 2,154-token router loads one skill per request out of 436,284 tokens of depth. 11 release-blocking gates, including one that checks the pack's own docs and reference material against its own rules.