Development tools and libraries for agent systems
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.
Local LLM runner: open-source models callable as a single CLI binary.
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.
llama.cpp is a lightweight C/C++ inference engine that runs a wide range of open-source large language models efficiently on consumer hardware.
OpenAI Whisper is a multilingual speech recognition foundation model that performs high-quality English and multilingual transcription, translation and language identification locally.
A high-throughput and memory-efficient inference and serving engine for LLMs, featuring PagedAttention, continuous batching, and optimized KV cache management for production deployments.
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.
Turn screenshots, mockups, and Figma designs into clean code using AI models. Supports HTML/Tailwind, React, Vue, and other frontend frameworks.
Daytona provides secure development-environment infrastructure for coding agents and automation workflows, serving as a runtime base for remote execution tasks.
A financial data platform for analysts, quants and AI agents, providing comprehensive financial data access across stocks, crypto, economics and more.
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.
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.
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.
GPT4Free (g4f) is a community-driven multi-provider LLM aggregation tool uniting dozens of LLM and media-generation APIs behind a single OpenAI-compatible interface. Includes Python/JS client, Docker images, and an MCP server.
Open Interpreter is a natural language interface for computers that lets LLMs run code locally to perform file operations, data analysis, and system management tasks.
AnythingLLM is an all-in-one AI productivity app with a self-hosted chat UI, RAG knowledge base, AI agents, and multi-model management, privacy-first with zero configuration.
Mem0 is a long-term memory layer for AI agents, supporting cross-session memory management and personalized context retrieval.
Context7 is Upstash's context-engineering toolkit for agents, helping applications manage long context windows, retrieval injection, and history compression.
LiteLLM provides a unified interface and proxy gateway for LLM calls, simplifying multi-model switching, routing, and cost control.
tldraw is an infinite canvas SDK for the web; its AI integrations let agents draw, annotate, and reason visually over a shared whiteboard.
Claude Cookbooks is Anthropic's official collection of notebooks and recipes showcasing fun and effective ways of using Claude, covering tool use, RAG, multimodal, and various agent application scenarios.
MCP server providing Chrome DevTools capabilities to coding agents, enabling web debugging, performance analysis, and DOM manipulation automation.
New API is a unified AI model hub for aggregation and distribution, supporting cross-conversion of various LLMs into OpenAI, Claude, or Gemini-compatible formats. A centralized gateway for personal and enterprise model management.
MindsDB is a query engine for AI analytics that enables building self-reasoning agents across live data, connecting diverse data sources with AI models.
(24 / 233)
A systematic comparison of the three categories of agent memory -- working, long-term, and shared -- covering storage media, lifecycle, retrieval methods, typical frameworks, and design patterns, fully addressing agent personalization and multi-agent collaboration engineering.
Based on OWASP LLM Top 10 engineering practice, this article systematically explains the seven layers of defense-in-depth for agent prompt injection: input sanitization, instruction isolation, least-privilege, output auditing, guardrails frameworks, continuous red-teaming, and kill switches -- with actionable code and toolchains.
Exploring how small language models are fine-tuned and deployed for agent workloads at the edge, balancing latency, cost, and accuracy for production AI agents.
Comparing container, WebAssembly, and process-level isolation approaches, with practical code for safely executing agent-generated code.
Breaking down three abstraction layers for browser automation—from raw Playwright to structured extraction—with production patterns, runnable code, and common pitfalls.
Learn how to build stateful AI agents with long-term memory using Letta (formerly MemGPT), solving the LLM context window limitation.