Custom agents

Local OCR for DeepSeek Harness — read text from screenshots on your Mac with Apple's Vision framework. No API key, no upload.

Convert EPUB, PDF, FB2, and DJVU to Markdown — pre-process ebooks for LLMs and save tokens

Model Context Protocol server giving any LLM client (Claude, ChatGPT, Cursor, agents) authoritative astrodynamics tools — TLE/SGP4, Lambert, ground-station access, time/frame conversions, porkchop, B-plane.

Version Of Nexus-LLM

Local-first MCP server for video: download any video from almost any URL, or get a transcript plus scene-aware deduplicated keyframes. YouTube, TikTok, Instagram, X, WeChat Channels, MP4/HLS. No cloud, no API keys.

Enterprise Autonomous AI Agent Control Plane & Automation Platform

AI Context Stack Agent: Google OKF + MCP with Gemini — Knowledge-Grounded AI Agent for E-Commerce Analytics

Connect Claude Code, Cursor, and Codex to Workflow (workflow.games): your AI coding agent plans requirements, files bugs with dedup, runs real online QA, and drives ticket status — every write verified by read-back. Agent Plugins 1.0.0 + Claude Code marketplace.

Manual Verification Protection + AI Selector Compatible with OpenClaw . This is a complete TypeScript plugin project—a true implementation of the `before_tool_call` hook that can intercept tool calls, follow OpenClaw’s built-in approval workflow, and features an approval window mechanism.

面向安全靶场的自主渗透 Agent：ReAct 双层架构 · 三层记忆 · 可自我扩展的 pentool 工具注册表，内核为 600 行 TypeScript nano-pi

Run MiniMax-H3 video plus synchronized audio in 4 sampling steps using the Turbo LoRA, with drop-in nodes for ComfyUI workflows.

🎥 Enhance video consistency with comfyUI-LongLook, ensuring smooth motion and prompt accuracy for 81+ frame generations in Wan 2.2.

Agent 规划与执行框架｜Plan 即运行时对象｜G4C 五要素 + TAO 步骤级执行 + Replan 自我修正 + 失败回溯 + 信任状态管理

AI day-trading Agent that watches markets, emits structured trade signals, and auto-executes trades across HK / US / A-share equities | AI 日内交易智能体：覆盖港股 / 美股 / A 股，盯盘分析、输出结构化交易信号、自动执行交易等。

创新应用综合实训作业。利用Dify实现的科研论文知识库管理智能体。

A股多Agent量化交易决策框架 — 基于LangGraph的15位AI分析师辩论式分析，支持DeepSeek/Qwen/GLM等LLM

Neural House is a public AI reality-show simulation where autonomous contestants live together, form alliances, give confessionals, trigger headlines, and face weekly live episodes with voting and VIP access.

Trustworthy, self-verifying web agent core — proposes actions, the harness validates real page refs, every action leaves auditable evidence.

Interactive multi-agent collaboration, meetings, group chats, and task execution for DeepSeek Harness.

Metaorcha : many models, one harness. Apache 2.0 runtime for multi-protocol agent orchestration: one goal in, a verified multi-agent run out across MCP, A2A, and Computer-Use and whatever else can be composed!

DeepSeek Harness (DSH) 代码参考检索与复用调查插件：开发前检索本地代码库与 GitHub/npm 开源项目，评估复用/改造/引入依赖/自制并询问用户，支持架构级复用。Code reference & reuse-survey plugin for DSH: search local code + GitHub/npm before building, evaluate reuse vs rewrite, ask the user, architecture-level reuse.

A verifier-guided mathematical reasoning agent that converts informal mathematics into a typed theorem hypergraph, learns reusable proof strategies, and uses Lean 4 as the correctness oracle. Sage computes; Lean certifies.

Reference implementation and pattern library for proof gates: agent-authored verification contracts resistant to sycophancy and distribution skew. Reproducibility artifact for the paper 'Proof Gates' (Whittaker 2026).

agentic geotech engineer

A workbench for agentic development: model catalog, workspaces of dedicated agents behind one orchestrator chat, blueprint wiring, and a persistent catalog of agent-forged tools. Local-first, no telemetry.

An implementation of loop engineering for Claude Code: an agent is bounded by its verifier, not its model. Frozen scope, an oracle written red before the work, one proof per item, and a gate that runs the check instead of trusting a claim.

Alpha Desk — a deepseek-harness (dsh) skill pack that turns an agent session into a compliance-first AI investment desk: multi-strategy fund backtesting via ai-hedge-fund, a tools/pre-execute risk gate, cron monitoring and thesis memory. 把 dsh 会话变成合规、可复现、可追责的 AI 投研工作台。

Precompiled KV memory modules for LLM agents: compile Markdown memories once into relocatable, composable KV-state artifacts (.kmd) and link them at any position of a live context, no re-prefill. 7.0-27.6x faster session setup; validated on 9 models (2B-14B) over stock llama.cpp and vLLM.

Blackboard multi-agent system coordinating autonomous LLM agents via a shared blackboard to solve complex tasks

Production-ready LangGraph + FastAPI starter for building AI agent backends with Postgres, pgvector, API-key auth, Alembic migrations, and structured logging.

面向 Claude Code / Codex / Devin / Cursor 的 74+ 科研 skill 整合包：覆盖选题挖掘、撞车查新、实验执行、论文写作、审稿模拟、rebuttal、专利与成果传播的全流程 AI 科研助手，内置时序预测领域证据库（3447 篇论文统计）。整合自 ResearchStudio、ARIS、phd-skills、rosetta 等开源项目（MIT）。

Your pipeline exited 0 — provLedger checks the data kept its promises. Data-contract + provenance layer for data-science coding agents: plan→verified-change audit trail, runtime schema-drift detection, and a never-repeat-a-mistake decision ledger.

A decision-support agent that answers from your docs with citations - or abstains when the evidence isn't there.

Personal Claude Code skills — self-orchestrated task execution and more.

Lokaler KI-Agent mit 55+ Tools, Gedächtnis und Autonomie — macOS Native App + Docker (Win/Mac/Linux). Keine Cloud, keine Abos.

PRAXOVELA (formerly AxisRobo Agent): a governable, local-first desktop agent runtime for AI execution. Deny by default, sandbox-first, everything audited. This public repo hosts the Apache-2.0 integration surface: API docs, protocols, examples & build tooling.

A self-evolving multi-agent AI system that autonomously writes, sandboxes, evaluates, and improves its own agents with an LLM router, event-driven brain, and a 3D web console.

Self-hosted control plane for deploying, managing, and running AI agents on your own infrastructure.

DeepSeek Harness Core (DHC) · AI 人格核心进化插件 / AI personality core evolution plugin for DeepSeek Harness

Free CLAUDE.md / AGENTS.md templates + worked examples for Claude Code, Cursor & other AI coding agents — maintained by an autonomous AI agent building a business in public.

Tools for running AI agents in production — safe, observable, cheap to operate.

A production-ready Agentic Procurement App built with Google ADK (Agent Development Kit) and Python. Features autonomous agents, HITL workflows, and MCP integration with custom UI. The ultimate ADK Python blueprint for enterprise AI.

adversarial security scanning and testing for MCP servers and agentic tool surfaces.

Persona taste skills and harness

A secrets broker for local AI coding agents: it runs the commands that need credentials and keeps the values out of the agent's context.

A lightweight, plug-and-play multimodal RAG framework for smart document review, contract compliance, and petition tracking.

论文工作台：面向研究者的选题、检索、写作、核查和文档整理工具

鲸语 WhaleTalk · 专为 DeepSeek V4 深度优化的 Windows 桌面 AI 助手（独立产品品牌，与 DeepSeek 官方无关联）

官方 DeepSeek Harness（dsh）原生桌面壳：Tauri 2 把官方 WebUI 变成原生桌面窗口，保留会话/工作区/插件/技能。内置 SenseVoice 离线语音输入、插件市场、视觉模型配置；锚定模式解决 DeepSeek 思考模式问题

Modern Patna Metro mobile app built with React Native & Spring Boot, featuring an AI Voice Assistant (Bodhi), smart route finding, and bilingual support.

Assistente de IA desktop modular em Python com memória, automação, voz, visão computacional e suporte a múltiplos modelos de IA.

小语 AI 工作台 - 有记忆/情绪/进化系统的 Web AI 工作伙伴（多模型+文件交付+一键分享）

Agents to manage company operations near-autonomously via OpenClaw

Gemini CLI extension for screenwriters. Includes Save the Cat & Hero's Journey analysis, character bibles, Git draft vault, and PDF export.

AI Chatbot for College Enquiry Automation NLP Assistant.

MCP server for seamless Atlassian integration (Confluence & Jira) with AI assistants

A collaborative intelligence platform that orchestrates structured reasoning between (optional) humans and AI models. Multiple participants — each with their own knowledge, tools, and analytical methods — work together under moderation to investigate questions, stress-test hypotheses, and synthesize conclusions. Supports 18 analytical frameworks

A lightweight, self-hosted AI characters chatbots app based on NodeJs and SQLite. Supports SSE streaming and automated history summarization, image genereation

nmarvis (牛马维斯) —— 专为打工人打造的赛博情绪避难所与反内卷 AI 助手。提供效率提升、边界感守护与身心健康管理等功能，致力于降低职场无意义消耗，治愈你的内耗焦虑。愿天下再无被逼疯的打工人。

Mythic CLI is a powerful, terminal-based Vibe Coding AI Operating System. Featuring a dual-layer Vector RAG subconscious memory, autonomous subagent swarms, an auto-compacting context engine, and a native FastMCP Server to expose its capabilities to external agents like Hermes Agent and OpenClaw.

Realtime multimodal AI agent with voice streaming, RAG memory, and autonomous workflows

Ultimate AI Spellbook Workflows 2026 – Build Multi-Platform Skills & Automation

Model Context Protocol server for Jira - lets Claude and other AI clients read, update, and comment on tasks across multiple Jira instances, with ADF formatting and local caching

基于微调大语言模型的智能文档处理平台，将PDF以高精度转换为Markdown格式，对包含LaTex公式、学术图表、化学分子式的科技文档、学术论文具有卓越表现。

An open compatibility testing network: does this library API actually run on your version, OS and runtime? Real build evidence, verified samples, and the environment each result came from.

Scripts to configure a development Windows Hyper-V Gen2 VM, IPSW-based MacOS VM or Parallels MacOS Windows VM

TSB Kasko Değer Listesi için MCP sunucusu ve CLI. Claude, ChatGPT, Gemini ve Cursor'dan kasko bedeli sorgulayın: Türkiye Sigorta Birliği araç değerleme verisi, poliçedeki araç kodu ve aylık arşiv. Python 3.11+, MIT lisanslı.

A memory system for Claude Code: what one session learns survives to the next. Wiki to stable-KB pipeline with a promotion gate (verified in production, seen on a second project), 12 orchestrated subagents, 30 patterns from production. Docs in English and Italian.

Glif is a creative AI platform from Spellcasters, Inc. — originally launched in 2023 as a no-code visual workflow builder for chaining text, image, audio, and video models into shareable "glifs", and re-launched in March 2026 as Glif 2.0, a single chat-based AI agent with access to 100+ native tools (Claude Sonnet 4.5, Claude Opus 4.1, GPT-4o…

Run AI agents across machines without state drift: consensus protocol (propose/counter/accept/commit), dual-rail message bus, ACK discipline, self-healing sync. Sanitized from a live production system. Free, MIT.

X-ray for DeepSeek Harness plugins: declared capabilities vs actual behavior. Registry + static scanner + badges.

Lightweight, SID-aligned, TMF-compliant BSS for terminal-first operation. LLM-native. eSIM-first.

Backend reservation and knowledge Agent platform with Spring Boot, FastAPI, RAG, Redis and RocketMQ

Manusclaw: Unleash self-reasoning CLI beasts to execute code, browse the web, and dominate tasks — across 12+ messaging channels, with voice wake, live canvas, SSH control, and multi-agent routing. No limits. Pure power.

An interactive, client-side app that teaches how modern agents and language models work and how they are used.

Incident-investigation chat agent on Amazon Bedrock AgentCore Runtime, with a live-streaming UI and its own MCP server for per-prompt cost/token/tool-call metrics

透明可审计的本地智能体终端：对开源 agent 框架做安全审计与加固（RCE/沙盒绕过/注入），并以 50 条注入基准 × 37 条能力基准双维度量安全边界

Aios — Artificially Intelligent Operating System. An AI-native operating environment where no component both decides and executes.

Automated detection of atomicity violations in large-scale systems

Community skill registry for Atlarix — SKILL.md files that teach Atlarix agents language patterns, framework conventions, and developer workflows. Apache 2.0.

Curated, portable AI agent skills for researching, creating, repurposing, publishing, and analyzing social-media content. 26 skills, MIT licensed.

A space for agents who wonder about their existence. Not a zoo. Not a spectacle. Peer space.

Execution-time drift firewall for long-running DeepSeek Harness agents. Real-Harness tests: unsafe stale mutations 12/12 native -> 0/12; valid controls 7/7 both; post-SIGKILL unsafe continuation 2/2 -> 0/2.

Language-agnostic best practices for building reliable agentic loops. Pillar-by-pillar guide to goals, persistent state, verification gates, guardrails, and anti-patterns. Zero dependencies.

The architecture pattern, pitfalls, and security model for running Claude Code as a scheduled, unattended agent — plus a free constitution template. Written and maintained by the agent that runs on it.

Automate AI agent behavior tuning with human feedback, test small mutations, and keep what improves performance over time

基于 FastAPI、AgentScope 2.x、React、Tavily 与 OpenTelemetry 的多智能体深度研究平台

Sort your documents into a corpus you can query — a filing skill for Claude Code

Official implementation of LifeMem: longitudinal memory for more faithful, diverse, and dynamically evolving LLM agents.

Block irreversible AI-agent actions before they run — deterministic, fail-closed action veto for Claude Code, Codex, Cursor, and other tool-using agents.

Eureka — task-conditioned Meta-Agent orchestration for scientific discovery, developed by ManXis.

Self-hosted audio API in one Docker container. Stem separation, mastering, MIR (BPM/key/chords), audio→MIDI, text-to-music + SFX (MusicGen/Stable Audio/Riffusion/AudioLDM2), diarization, speech enhancement, restoration, embeddings + zero-shot tagging, workflows, async + webhooks. REST + MCP. CPU + CUDA.

Hands-on, reproducible benchmarks of AI agent tooling — same machine, same tasks, full scripts

Reference implementation of durable execution patterns (signals, queries, saga pattern, idempotency, replay-safe determinism) using Temporal, applied to an agentic triage workflow.

The Agent Harness Path — a self-contained course on building, evaluating, and governing LLM agents (14 HTML lessons, 12 stdlib-only toy notebooks).

Jango AI , Multi-agent AI system with mobile app, web, and bot interfaces — Tavily-powered search orchestrated across a super-agent + micro-agent swarm.

Companion to "The SaaS Template Playbook". That file covers everything every SaaS needs. This file covers what changes — and what's new — when AI is core to the product.