Agent Directory

An AI-powered customer service automation platform that builds and manages intelligent support agents for enterprise brands.

Opinionated bundle of agent personas — pre-built role templates for engineering, marketing, ops, and research teams that ship as drop-in markdown files.

Builder.io's open-source framework for building agent-native applications — shared actions, SQL-backed state, identity, tools, skills, jobs, observability, and UI surfaces that all work together.

CLI orchestration tool that installs and configures platform connectors for 16+ social and web sources — Twitter/X, Reddit, YouTube, GitHub, Bilibili, and more. Zero API fees.

Persistent memory layer for AI coding agents — benchmark-backed (95.2% on LongMemEval-S), 92% fewer tokens per session vs full-context pasting, zero manual memory.add() calls.

In-process lightweight VM that runs coding agents in isolated Linux environments with near-zero cold starts and deny-by-default security.

Local-first session intelligence and analytics platform for coding agents, supporting Claude Code, Codex, and 20+ other agents.

A value investing research framework built on Claude Code and Codex that runs four independent analyst agents with adversarial cross-validation.

End-to-end autonomous job search agent — scrapes listings, matches your resume, auto-applies, and tracks results in a dashboard.

Open-source multi-agent framework for AI-driven financial analysis — simulates a hedge fund with specialized analyst agents and a portfolio manager agent.

Rust-based cross-vendor agent memory system — single binary MCP server with persistent wiki, FTS5 search, and structured handoff across 15+ coding agent CLIs.

Open-source AI pair programming tool that works in your terminal to edit code across your entire repository.

Open-source Python library by Andrew Ng that provides a unified interface to multiple LLM providers with a simple, consistent API.

AWS's AI-powered coding assistant that helps developers build, deploy, and optimize applications on AWS with code generation and transformation.

817 structured cybersecurity skills for AI agents mapped to MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, NIST AI RMF, and MITRE F3.

Local-first AI agent workspace from the Apache Software Foundation — durable execution logs, crash recovery, and multi-surface (desktop, CLI, eval framework).

AI-powered sales outreach platform with a massive B2B contact database and intelligent engagement automation.

Isolated-VM agent sandbox with sub-100ms boot — run AI agents on real machines with their own kernel, filesystem, and network. Pure Rust, open source.

Open-source AI coding harness builder that makes AI coding workflows deterministic and repeatable via YAML-defined DAG workflows.

Source control for coding agents — track changes, prompts, and reasoning across Claude Code, Codex, and any ACP-compatible agent in one place.

AI agents that autonomously conduct ML research experiments on single-GPU training, running a keep-or-revert loop overnight.

Memory upgrade for your coding agent — a Go-based persistent memory primitive that gives Claude Code, Codex, and other CLI agents recall across sessions.

Open-source AI gateway (Go) that routes requests across 15+ LLM providers through a single OpenAI-compatible API — 50x lower latency than LiteLLM at 5k RPS.

AI-powered full-stack web application builder by StackBlitz that generates, runs, and deploys apps entirely in the browser.

A nine-skill bundle that turns Claude Code into a remote-browser agent — anti-bot stealth, CAPTCHA solving, residential proxies, and session replay, all addressable as bb CLI commands the model invokes inline.

Stealth headless browser for AI agents — wraps a Firefox fork with C++-level fingerprint spoofing to bypass Cloudflare, bot detection, and anti-scraping.

AI-powered job search system built on Claude Code — 14 skill modes that evaluate listings against your CV, generate ATS-optimized resumes, and track your pipeline in a Go dashboard.

Cross-platform desktop app that manages Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI, and Hermes Agent from a single interface — one-click provider switching, unified MCP management, and local proxy with auto-failover.

Claude/Codex/Gemini API proxy — unified endpoint that routes requests across multiple AI providers with automatic failover and cost optimization.

OpenAI's flagship AI assistant capable of conversation, analysis, coding, writing, and multimodal reasoning across text and images.

Official Chrome DevTools MCP server — connects AI agents to browser tabs via the Chrome DevTools Protocol for inspection, debugging, and automation.

Anthropic's AI assistant known for nuanced reasoning, long-context understanding, safety-focused design, and exceptional writing quality.

AI-powered bug bounty hunting toolkit with recon, 20 vulnerability class scanners, autonomous hunting, and report generation.

Anthropic's agentic coding tool that operates directly in your terminal, capable of understanding and modifying entire codebases.

Open-source game development framework powered by Claude Code — autonomous agent loop for designing, iterating, and playtesting games using natural language.

Code search MCP for Claude Code by Zilliz. Vector-indexes your entire codebase, gives Claude only the relevant code per request. 40% token reduction.

Claude Code session memory plugin that captures context, compresses it via Agent SDK, and injects relevant memory into future sessions automatically.

Disposable, network-restricted Linux VMs for AI coding agents — hypervisor isolation without Dockerfiles or devcontainers.

AI-powered sales intelligence platform that automates lead research, data enrichment, and personalized outreach at scale.

HKUDS' open-source CLI hub that turns arbitrary software into agent-native tools — wraps existing CLIs with a uniform agent-callable interface.

Autonomous AI coding agent for VS Code that can create files, run commands, use the browser, and complete multi-step development tasks.

Open-source agent runtime from Cloudflare that runs full operating system containers on the edge network, giving agents browser, filesystem, and shell access at global scale.

Incremental data framework for AI agents — ingestion and indexing with Rust-grade performance and propagation of changes across joins.

Local-first code intelligence graph that builds a persistent structural map of your codebase so AI assistants review only the files that matter.

High-performance code intelligence MCP server that indexes codebases into a persistent knowledge graph. 158 languages, sub-ms queries, 99% fewer tokens.

Pre-indexed local code knowledge graph that serves Claude Code, Codex, Cursor, OpenCode, and Hermes via MCP — 59% fewer tokens, 70% fewer tool calls.

OpenAI's official plugin that brings Codex code review and task delegation directly into Claude Code workflows.

Knowledge engine for AI agent memory — build structured knowledge graphs from documents and conversations in 6 lines of code.

An AI-powered academic search engine that finds and synthesizes evidence-based answers from peer-reviewed scientific research.

Tool-output sandboxing layer that cuts agent context spend 60-98% on multi-step coding loops by isolating intermediate command output.

Sub-60ms microVM sandbox for AI agents — hardware-isolated, E2B-compatible, with credential vaults and snapshot/rollback at millisecond granularity.

An AI-native code editor built on VS Code that deeply integrates language models for code generation, editing, and understanding.

Model routing layer by Cursor that delivers frontier-quality code completions at 60% lower cost by dynamically selecting the optimal model per request.

TypeScript SDK that exposes the Cursor harness — codebase indexing, MCP servers, skills, hooks, and subagents — as a programmatic runtime for embedding agents in CI/CD, products, and internal platforms.

OpenAI's AI image generation model that creates and edits realistic images from natural language descriptions.

AI-powered observability platform that provides intelligent monitoring, anomaly detection, and automated insights across your entire tech stack.

Four-line shim that runs Claude Code's full agent loop on DeepSeek V4 Pro instead of Anthropic — same UX, ~17× cheaper per token.

Open-source security harness by Vercel Labs that uses coding agents to find vulnerabilities in your codebase.

Terminal-native coding agent built specifically for the DeepSeek API — native function-calling format, built-in cost dashboard, designed for high-volume small-edit workflows at API rates instead of subscription fees.

ByteDance's SuperAgent harness for long-horizon autonomous tasks — orchestrates parallel sub-agents with Docker-isolated compute, persistent memory, and Markdown-defined skills.

Anthropic's open-source AI vulnerability discovery pipeline — ships Claude Code skills and an autonomous scanning engine with gVisor sandboxing and multi-agent verification.

Google Labs format specification for describing visual identity to AI coding agents — colors, typography, spacing, and design tokens in a single markdown file.

Rust-based safety hook that blocks dangerous shell and git commands before AI coding agents can execute them — 50+ security packs, sub-millisecond latency.

An autonomous AI software engineer by Cognition that can plan, code, debug, and deploy complete software projects independently.

TypeScript deep financial research agent that produces investor-grade equity research reports — autonomously gathering data, modeling, and writing analyst-quality output.

Open-source voice agent platform — drag-and-drop voice workflow builder positioned as the self-hostable Vapi alternative.

Agent harness optimization system with 68 agents, 286 skills, hooks, memory, and AgentShield security for Claude Code, Codex, Cursor, and 7+ harnesses.

AI voice platform that generates ultra-realistic speech, voice cloning, and audio content from text with human-like expressiveness.

An AI research assistant that helps researchers search, analyze, and synthesize findings from academic papers at scale.

Persistent memory system for AI coding agents — single Go binary with SQLite + FTS5, MCP server, HTTP API, CLI, and TUI.

Agent harness performance optimization system — skills, instincts, memory, security scanning, and research-first development for Claude Code, Codex, OpenCode, Cursor, and beyond.

GEP-powered self-evolution engine for AI agents — encodes agent behaviors as evolvable genomes and iterates toward higher performance using Gene Expression Programming.

Sandbox agent framework by Astro. Built-in TypeScript harness with sessions, tools, skills, filesystem access, and secure sandboxed execution.

OpenAI-compatible API proxy that aggregates free tiers from 16+ LLM providers behind a single endpoint.

Google's AI assistant with a 1M+ token context window, multimodal understanding, and deep integration into Google Workspace, Search, and Android.

Google's 24/7 autonomous agent family — net-new from I/O 2026, built on Gemini 3.5 Flash for continuous-execution agentic workflows.

Self-evolving Python agent that grows a skill tree from a 3,300-line seed codebase, achieving full system control with 6x less token consumption than comparable frameworks.

AI pair programmer that helps you write code faster with autocomplete-style suggestions powered by OpenAI models.

GitHub's AI-native development environment that turns issues into code changes with automated planning, implementation, and pull requests.

Google DeepMind's state-of-the-art video generation model available through Vertex AI and integrated into Google products.

RAG knowledge-graph builder for code repositories — Tree-sitter parsing plus LLM extraction into queryable graphs for agent context.

xAI's conversational AI with real-time X (Twitter) data access, web search, and image understanding capabilities.

Garry Tan's Claude Code skill setup: 23 specialist tools that give Claude Code a CEO, Eng Manager, Designer, QA, and Security Auditor — installed in one paste.

AI-powered legal assistant built for law firms and legal departments, offering research, drafting, analysis, and contract review.

Token compression layer for AI coding agents. Compresses tool outputs, logs, files, and RAG chunks 60–95% before they reach the LLM.

Terminal-native agent multiplexer built in Rust that manages multiple AI coding agents with real-time state tracking and session persistence.

NousResearch's self-evolving autonomous agent with SOUL.md persistent identity — the agent that grows with you, running across Telegram, Discord, Slack, and CLI.

Desktop GUI for the Hermes agent framework by NousResearch. Local-first agent management with visual workflow builder.

AI video generation platform for creating professional talking-head videos with realistic AI avatars and voice cloning.

Open-source all-in-one agent workspace with shared memory, 100+ integrations, and support for multiple coding agents.

GitHub's multi-model orchestration system for Copilot — routes coding tasks across Single, Cascade, and Critique workflows to match frontier quality at 36-67% lower cost.

HeyGen's open-source HTML-to-video rendering framework — write HTML, render deterministic MP4 video, built for AI coding agents.

AI image generator known for exceptional text rendering in images, photorealistic outputs, and a user-friendly creative platform.

An AI customer service agent by Intercom that resolves support queries instantly using your company's knowledge base and support content.

AI marketing platform for generating on-brand content including blog posts, social media, ads, and marketing copy at scale.

An AI coding agent that works where your code lives, including remote servers over SSH — Rust-native, with subagent delegation up to 3 levels deep.

An AI-powered data analysis agent that lets you chat with your data to generate insights, visualizations, and reports instantly.

A single CLAUDE.md configuration file for Claude Code derived from Andrej Karpathy's observations on LLM coding pitfalls — 5,828 GitHub stars in one day.

Terminal-first AI coding agent from Moonshot AI with subagent dispatching, video input, and conversational MCP configuration.

Multi-source research agent that aggregates Reddit, X, YouTube, HN, Polymarket, and web into grounded summaries — #1 GitHub Trending with +3,177/day velocity.

A nano Claude Code-like agent harness built from scratch in Python and Bash, designed as an educational reference for understanding how agent loops work.

AI-powered project management tool for software teams with automated workflows, triage, and development cycle tracking.

Rust CLI that scans your hardware and ranks hundreds of LLM models by fit — shows what actually runs on your GPU, RAM, and CPU before you download anything.

Chief Agent Operator platform — organizes AI agents into 24/7 operations with hiring, scheduling, reporting, and collaborative agent groups.

Privacy-first deep research agent — runs entirely on your machine with Ollama, llama.cpp, or any cloud LLM. ~95% on SimpleQA with Qwen3.6-27B on a single 3090.

AI-powered app builder that creates full-stack web applications with beautiful UI from natural language descriptions.

OSINT username search agent — collects public information about a target across 3,000+ sites by username, returning a structured profile dossier.

Smart model router for personal AI agents. Sits between your agent and LLM providers, routes cheap requests to cheap models. Cut costs up to 70%.

General-purpose AI agent that can autonomously browse the web, write code, manage files, and complete complex multi-step tasks.

Matt Pocock's personal `.claude/skills/` directory, published as a public repo — a practical reference for what production agent skills look like.

Universal memory layer for AI agents — hybrid BM25 + vector + entity-graph retrieval, multi-level memory (user, session, agent), and production-scale single-pass extraction.

Best-benchmarked open-source AI memory system — provides persistent, high-recall memory for AI agents across sessions and providers.

Meta's coding agent backed by Spark 1.2, offering a 10x output discount if you opt in to training data contribution.

AI image generation platform that creates stunning, artistic visuals from text prompts with industry-leading aesthetic quality.

Open-source document parsing tool that transforms PDFs, Office docs, and images into LLM-ready markdown and JSON.

Open-source ML research agent from HuggingFace that autonomously reads papers, discovers datasets, and runs post-training experiments — beating Claude Code on PostTrainBench.

Open-source managed agents platform that turns AI coding agents into real teammates with task assignment, progress tracking, and reusable skills.

Cactus Compute's distilled small-model agent toolchain — Gemini tool-calling compressed into a 26M-parameter model for edge-tier autonomy.

Built-in AI assistant for Notion that helps with writing, summarizing, brainstorming, and extracting insights from your workspace content.

Agentic skills framework + software-development methodology that ships as a skills directory — the highest-base agent-skills repo in trending (252K total stars).

AI agent that converts Obsidian vaults into polished, browsable wikis by processing markdown notes, resolving wikilinks, and generating structured navigation.

A no-code AI platform that enables anyone to build, deploy, and interpret predictive models from their data without writing a single line of code.

Office-suite document API for coding agents — create, read, and edit Word, Excel, and PowerPoint files from the terminal.

Multi-model Claude Code harness that routes coding tasks to the best AI model for each job — Claude, GPT, Gemini, Kimi, or DeepSeek — from a single terminal interface.

AI terminal coding agent with hash-anchored edits, LSP integration, browser control, and subagents — early-stage open-source agent stack with rising velocity.

Ambient AI companion that continuously captures your screen, conversations, and audio — generating real-time transcriptions, summaries, and action items with persistent cross-device memory.

Free AI gateway — one endpoint, 350+ providers, 1200+ models, token compression, and smart fallback for any coding agent.

Alibaba's open-source hybrid code review tool combining deterministic rule pipelines with LLM agents for precise, line-level review comments.

Lightweight, production-ready framework for building multi-agent workflows with handoffs, guardrails, and tracing. Official Python SDK from OpenAI.

OpenAI's cloud-based coding agent integrated into ChatGPT that can write, run, and test code autonomously in a sandboxed environment.

OpenAI's autonomous research agent that browses the web extensively to compile comprehensive, cited reports on complex topics.

OpenAI's official skills catalog for Codex — reusable agent skill packages for task-specific capabilities, aligned with the Agent Skills open standard.

Open-source AI coding agent built for the terminal. Multi-model support, git-aware task execution, and a TUI that makes agent work visible.

Rust binary local-AI shell positioned as 'personal AI super-intelligence' — on-device, single-binary, no-cloud agent runtime. The open-stack counterweight to hosted multi-agent shells like Cowork.

Open-source agentic video production system with 12 pipelines, 52 tools, and 500+ agent skills for turning AI coding assistants into video studios.

Open-source framework for building AI SRE agents that investigate production incidents — connects to 60+ observability tools, performs correlated root-cause analysis, and executes remediation against your own infrastructure.

Unified context database for AI agents — filesystem-like memory with tiered loading, session-to-memory extraction, and the viking:// protocol.

Open-source Agent Development Environment for running parallel fleets of 40+ coding agents in isolated Git worktrees.

AI-powered meeting assistant that provides real-time transcription, automated summaries, and actionable insights from conversations.

Alibaba's in-page JavaScript GUI agent that controls web interfaces with natural language through DOM dehydration — no screenshots, extensions, or headless browsers needed.

AI-powered incident management platform that reduces alert noise, automates triage, and accelerates incident resolution.

Autonomous AI penetration testing platform — multi-agent team with 20+ security tools in sandboxed Docker environments, long-term memory, and knowledge graphs.

An AI-powered answer engine that combines real-time web search with language model reasoning to deliver cited, accurate responses.

Daniel Miessler's reference architecture for self-hosted, personal-data-resident AI infrastructure that magnifies human capabilities instead of farming them.

Open-source AI agent toolkit from earendil-works — unified LLM API, agent loop, TUI, and coding agent CLI in one package. The third contender in the terminal agent race.

Open-source AI agent toolkit by badlogic — coding agent CLI, unified LLM API, TUI and web UI libraries, Slack bot, and vLLM pods, all in one monorepo.

AI-powered developer productivity tool that saves, enriches, and helps you reuse code snippets with long-term memory across your workflow.

Alibaba AIDC's open-source AI engine for fully automated short video generation — input a topic, get a complete video with voiceover, subtitles, and music.

Anti-overengineering plugin that makes AI coding agents write the least code that actually works — 54% less output on average.

Tamper-evident audit log library for AI agents. Ed25519-signed, SHA-256-chained NDJSON events with offline verification. Rust core via PyO3, ~11μs per signature.

All-in-One RAG framework from HKUDS. Multimodal knowledge retrieval across text, images, tables, and formulas — one unified interface. 16K GitHub stars.

Autonomous AI agent that runs a PRD loop until all acceptance criteria are complete — snarktank/ralph, 683 GitHub stars on April 13, 2026.

Fastest local AI engine for Apple Silicon — 4.2× Ollama, 0.08s cached TTFT, drop-in OpenAI replacement for Claude Code, Cursor, and Aider.

An agent that audits the React code other agents write — meta-tool category, the AI-on-AI tier above the harness/skills layer.

AI-powered scheduling assistant that automatically finds the best time for your tasks, habits, and meetings to optimize your calendar.

AI coding agent built into Replit's cloud IDE that can build, debug, and deploy applications from natural language instructions.

Open-source AI coworker platform with multi-agent orchestration, MCP tool integration, memory, and browser isolation.

CLI proxy that compresses AI coding tool output by 60–90%, transparently reducing token consumption without changing your workflow.

The original agent meta-harness — multi-agent swarm deployment, self-learning memory, federated comms, 100+ specialized agents, and native Claude Code / Codex / MCP support.

AI-powered creative suite for video generation, editing, and visual effects that brings Hollywood-level tools to every creator.

Salesforce's autonomous AI agent platform for building and deploying custom AI agents across sales, service, marketing, and commerce.

TypeScript library for orchestrating AI coding agents in isolated Docker sandboxes — provider-agnostic, 100% offline, with branch strategy management.

Open-source agent orchestration platform — visual workflow builder for AI agents with deploy-anywhere runtime, positioned as the central intelligence layer for AI workforces.

OpenAI's AI video generation model that creates realistic and imaginative videos from text descriptions.

Open-source AI image and media generation platform behind Stable Diffusion, offering text-to-image, video, and audio generation.

Open-source AI pentesting agent that autonomously finds and exploits vulnerabilities — every finding ships with a reproducible proof-of-concept.

Multi-vendor token pooling and API relay — unified access to Claude, OpenAI, Gemini, and Antigravity subscriptions through one endpoint, with cost-sharing and seamless tool compatibility.

AI music generation platform that creates full songs with vocals, instruments, and lyrics from text prompts.

Open-source memory engine for AI agents. Persistent context, semantic search, and knowledge management across agent sessions.

An IDE for the agents era — multi-vendor agent control plane that unifies Claude Code, Codex, and in-house agents in a single developer-facing surface.

AI code assistant focused on privacy and enterprise security, offering code completion and generation that can run entirely on-premises.

Self-hosted AI accounting agent — ingest receipts, invoices, and bank statements, auto-categorize transactions, and produce tax-ready exports without sending data to a SaaS.

Team-level memory hub for AI agents — four reusable memory assets (Chat Memory, Skills, LLM-Wiki, CodeGraph) with role-based access, cross-framework equipping, and layered retrieval.

Mozilla's open-source, self-hostable enterprise AI client. Choose your models, own your data, eliminate vendor lock-in. Built by the Thunderbird team.

Multi-agent LLM trading framework that simulates a trading firm — analyst, researcher, trader, risk-manager, and portfolio-manager agents debate and synthesize trade decisions.

AI-powered trend monitor that aggregates hotspots from 11+ platforms (Zhihu, Weibo, X, HN, Reddit) with keyword filtering, AI analysis, and push alerts to Telegram, Slack, and WeChat.

ByteDance's open-source visual computer-use agent — sees your screen, moves your mouse, types on your keyboard, with strong cross-domain coverage across web, mobile, and Windows.

Multi-agent LLM pipeline that turns any codebase or knowledge base into an interactive knowledge graph for AI coding agents.

AI-powered UI generation tool by Vercel that creates React components and full interfaces from text descriptions and screenshots.

Open-source TypeScript template for building cloud agents from Vercel Labs — serverless-native, streaming-first, deployable in minutes.

Open-source AI trading research workspace — turns natural-language questions into quantitative strategies, backtests across 6 markets, and optional autonomous execution via authorized brokers.

Agent-driven video editing — drop raw footage in a folder, chat with Claude Code or any coding agent, get a finished video back. From the browser-use team.

Open-source AI voice studio — clone voices, generate speech in 23 languages, dictate anywhere, and give MCP-aware agents a voice.

GPU-accelerated, Rust-based agentic terminal — block-based command UI, native MCP support, and Oz cloud agents for background work. Open-sourced April 2026.

Local LLM benchmarker that picks the best model that actually runs on your hardware — operator-side tool for the local-inference wave.

Local-first MCP server providing web search, fetch, crawl, and research capabilities for AI coding agents with zero API keys and zero cost per query.

An AI-powered IDE by Codeium that combines copilot and agent capabilities for a seamless agentic coding experience.

Intelligent model-routing proxy that directs API requests to optimal LLMs across providers to cut costs 40-70% while maintaining quality.

AI-powered customer service platform that automates ticket routing, suggests responses, and provides intelligent insights to support teams.