Agent Tools

assistantmulti-channelllm +2

OpenClaw is an open-source personal AI assistant platform supporting 25+ messaging channels (WhatsApp, Telegram, Slack, etc.) with multi-LLM integration and personal knowledge management.

· 25+ messaging channels - Unified access to WhatsApp, Telegram, Slack, Discord, WeChat, QQ, and more
· Multi-LLM support - Works with OpenAI, Anthropic, and other providers via OAuth or BYOK
· Voice interaction - Speech input/output on macOS, iOS, and Android for hands-free conversation

ECC

240.5k · JavaScript

agent-harnessskillsclaude-code +2

Agent harness operating system providing skills, instincts, memory optimization, continuous learning, and security scanning.

· Cross-harness workflows - supports Codex, Claude Code, Cursor, OpenCode, Gemini and more
· Skills system - reusable agent skill packs covering code generation, review, and testing
· Continuous learning - mines failed sessions for lessons, auto-writes corrections

Ollama

178.7k · Go

Local LLM runner: open-source models callable as a single CLI binary.

llmlocalinference +1

· One-command run — `ollama run llama3` drops into a chat
· Model registry — Llama, Mistral, Qwen, Gemma built-in
· OpenAI compatible — Serves /v1/chat/completions

Firecrawl

168.2k · TypeScript

web-scrapingsearch-enginemarkdown +2

Firecrawl is a web scraping and search engine designed for AI agents, converting any webpage into structured Markdown data with search, scrape, and clean capabilities for building web-data-powered AI applications.

· Search endpoint — Search the web and get full page content from results in a single call
· Scrape endpoint — Convert any URL to clean Markdown, HTML, screenshots, or structured JSON
· Interact endpoint — Scrape a page then interact via AI prompts or code (click, search, navigate)

Excalidraw

129.8k · TypeScript

Open-source hand-drawn whiteboard tool. Combines with AI for diagram / flowchart generation, often used in Agent UI display.

whiteboardvisualdiagram

llama.cpp

124.2k · C++

llm-inferencellamagguf +2

llama.cpp is a lightweight C/C++ inference engine that runs a wide range of open-source large language models efficiently on consumer hardware.

· Ultra-lightweight inference — pure C/C++ with zero dependencies, runs quantized LLMs on CPU
· GGUF format — unified quantized model format that is cross-platform and supports partial loading
· Hardware acceleration — Apple Silicon Metal, NVIDIA CUDA, AMD ROCm, Vulkan and OpenCL backends

Whisper

107.4k · Python

speech-to-textasraudio +2

OpenAI Whisper is a multilingual speech recognition foundation model that performs high-quality English and multilingual transcription, translation and language identification locally.

· Multilingual — transcription across 99 languages plus translation to English
· Multiple sizes — from tiny to large, pick the trade-off between accuracy and speed
· Robust — handles accents, background noise and other real-world audio conditions

Ponytail

104.1k · JavaScript

agent-skillscode-qualityYAGNI +2

Skill pack that makes your AI agent write minimal code like a lazy senior dev, cutting code by 54% and cost by 20%.

· Minimalist code philosophy - guides agents to prefer native APIs over dependencies
· Measurable results - 54% less code on average, up to 94% on real projects
· Safety guards - retains all safety checks without sacrificing quality for brevity

Pi Mono

91.5k · TypeScript

agent-toolkitclillm-api +1

Pi Mono is a comprehensive AI agent toolkit including a coding agent CLI, unified LLM API, TUI and web UI libraries, Slack bot, and vLLM pod management for end-to-end agent development.

· Interactive coding agent CLI — terminal-based AI coding assistant for code generation, refactoring, and debugging
· Unified multi-provider LLM API — single interface supporting OpenAI, Anthropic, Google and other LLM providers
· Agent runtime engine — built-in tool calling and state management for complex multi-step task execution

Pi

91.5k · TypeScript

Pi is a full-stack AI agent toolkit featuring a coding agent CLI, unified LLM API, TUI and Web UI libraries, Slack bot, and vLLM inference pods.

coding-agentclillm +2

· Unified multi-provider LLM API — pi-ai package provides unified interface for OpenAI, Anthropic, Google, and other LLM providers
· Interactive coding agent CLI — pi-coding-agent provides command-line interactive coding agent with tool calling and state management
· Agent runtime — pi-agent-core provides agent runtime with tool calling and state management for multi-step task execution

vLLM

89.2k · Python

A high-throughput and memory-efficient inference and serving engine for LLMs, featuring PagedAttention, continuous batching, and optimized KV cache management for production deployments.

llmpythonframework +2

· PagedAttention memory management - Efficiently manages attention KV cache via paging, dramatically reducing memory usage
· Continuous batching and chunked prefill - Supports continuous batching, chunked prefill, and prefix caching for high-throughput inference
· Multi-quantization support - Covers FP8, INT8, INT4, GPTQ/AWQ, GGUF, compressed-tensors, and more quantization formats

PaddleOCR

87.7k · Python

ocrpaddlepaddledocument-ai +2

PaddleOCR is Baidu's open-source multilingual OCR and document-AI toolkit supporting 80+ languages and PP-Structure layout analysis.

· Multilingual recognition — 80+ languages out of the box including Chinese (Simplified/Traditional), English, Arabic
· PP-Structure layout analysis — text blocks, tables, figures, formulas auto-classified and located
· PP-OCRv4 high accuracy — state-of-the-art detection + recognition end-to-end accuracy

Open Design

87.5k · TypeScript

open-design is an AI co-design platform for designers and developers, unifying Figma, drawing and code workflows into agent tasks.

designfigmaai +2

· AI design agent — natural-language prompts generate Figma components, pages and interactions
· Bi-directional Figma sync — every AI edit is written back to the source Figma file in real time
· Component-library aware — auto-detects the project's existing design system to keep style consistent

Odysseus

85.5k · Python

browserautomationai-agent +2

odysseus is a macOS-native AI browser automation agent that drives Chrome with natural language to complete everyday workflows.

· macOS-native integration — deep hookup with system clipboard, shortcuts and notifications
· Chrome automation — drives Chrome via CDP following least-privilege principles
· Natural-language tasks — describe tasks and the agent decomposes and executes them

WorldMonitor

82.4k · TypeScript

monitoringworld-newsai +2

worldmonitor is an AI global event monitoring dashboard aggregating news, geopolitics, energy and market data with auto-generated briefings.

· Global news aggregation — multi-language sources, geopolitical events, market sentiment
· Event correlation — auto-links news events to energy, shipping and currency prices
· AI briefings — auto-generated daily / weekly briefings you can subscribe to

Netdata

80.2k · Go

monitoringmetricsreal-time +2

Netdata is an open-source real-time observability platform with second-resolution metrics, AI health checks and zero-config deployment, widely used for monitoring AI agent services.

· Real-time metrics — per-second collection of system, app, container and Kubernetes metrics
· AI anomaly detection — learns baselines and alerts on outliers automatically
· Zero-config deploy — one command starts an agent with dashboards, no extra config

DeerFlow

80.1k · Python

super-agentmulti-agentresearch +4

An open-source long-horizon SuperAgent harness by ByteDance that researches, codes, and creates with sandboxes, memories, tools, skills, subagents and message gateway for complex tasks.

· Super Agent harness - orchestrates sub-agents, memory and sandboxes for complex long-horizon tasks from minutes to hours
· Extensible skills system - modular skill-based architecture for extending agent capabilities with custom skills
· Sandboxed code execution - built-in secure sandbox for agents to safely execute code and file operations in isolation

Paperclip

78.5k · TypeScript

ai-agentsorchestratorscheduler +2

paperclip is an AI agent team orchestration and scheduling platform, managing task assignment, collaboration and state across many agents.

· Multi-agent orchestration — schedules many AI agents to complete complex tasks in parallel
· Task assignment — auto-matches tasks to agents based on capability
· Status tracking — live dashboard of each agent's state, progress and deliverables

Taste-Skill

77.1k · JavaScript

agent-skillscode-qualitywriting-style +2

Skill that gives your AI good taste, stopping agents from generating boring, generic slop content.

· Anti-AI-slop - prevents agents from generating boilerplate, generic content
· Taste guidance - trains agents to write distinctive, concise, direct content
· Cross-agent - loadable as a skill in Claude Code, Cursor and more

AppFlowy

75.6k · Dart

note-takingproject-managementai +2

AppFlowy is an open-source Notion alternative with AI workflows, knowledge base, kanban and multi-device sync, usable as agent long-term memory.

· Self-hostable — full data ownership, deploys locally or on a private server
· AI assistant — built-in AI for writing, summarizing, translating and table generation
· Knowledge base — nested pages, databases, linked views for structured notes

Screenshot to Code

74.0k · Python

screenshot-to-codemultimodalcode-generation +2

Turn screenshots, mockups, and Figma designs into clean code using AI models. Supports HTML/Tailwind, React, Vue, and other frontend frameworks.

· Multi-format input - converts screenshots, UI mockups, Figma designs and screen recordings into runnable code
· Multi-stack output - generates code for HTML+Tailwind, React+Tailwind, Vue+Tailwind, Bootstrap, Ionic and more
· Multi-model AI integration - built-in Gemini, GPT-5 series, Claude Opus models with side-by-side comparison

Front-End Checklist

73.5k · MDX

front-endchecklistweb-development +2

A comprehensive front-end checklist for modern web development and AI agents, covering HTML/CSS/JS, performance, accessibility, and SEO quality standards.

· Complete quality checklist - covers HTML, CSS, JavaScript, images, fonts, performance, accessibility, SEO, security and 20+ other categories
· AI-agent friendly - each item has clear check prompts that can be directly fed to LLMs for automated quality inspection
· Multi-language support - provides Chinese, English, Portuguese and other language versions

Unsloth

72.6k · Python

model-trainingfine-tuningllm +2

Unsloth Studio is a web UI for training and running open models like Gemma 4, Qwen3.6, DeepSeek, and gpt-oss locally, providing model fine-tuning and deployment capabilities for agent developers.

· High-efficiency fine-tuning engine - Custom Triton kernels for 2x faster training with 70% less VRAM, no accuracy loss
· Multi-modal training - Supports fine-tuning text, vision, TTS, and embedding models
· Reinforcement learning (RL) - GRPO, FP8 and other RL methods with 80% less VRAM usage

Daytona

72.0k · Unknown

dev-environmentsandboxcoding +1

Daytona provides secure development-environment infrastructure for coding agents and automation workflows, serving as a runtime base for remote execution tasks.

· Sandboxes - fully isolated compute units with dedicated kernel, filesystem, network stack, and allocated vCPU/RAM/disk
· Sub-90ms startup - sandboxes spin up in under 90ms from code to execution, supporting Python/TypeScript/JavaScript
· Multi-language SDK - Python, TypeScript, Go SDKs plus REST API and CLI for programmatic control