Catalogue

Agenta

Agenta-AI

Evaluate and compare different agent configurations side by side.

AgentOps

Monitor and benchmark agent performance in production.

aider

Aider-AI

Terminal-based AI pair programming, git-aware and ready in your shell.

Browser Use

browser-use

Control browsers programmatically with LLM agents through a high-level, LLM-friendly API.

Cavegemma

JuliusBrussee

An experimental LoRA fine-tune of Gemma to speak caveman-mode natively.

Cavekit

JuliusBrussee

A spec-driven build framework using caveman encoding for roughly 75% fewer tokens.

Caveman

JuliusBrussee

A Claude Code skill that cuts roughly 65% of output tokens via caveman-speak.

caveman-code

JuliusBrussee

A TypeScript implementation of the caveman compression engine.

Cavemem

JuliusBrussee

Cross-agent persistent memory, caveman-compressed into SQLite with FTS5 and served via MCP.

Chatterbox

resemble-ai

State-of-the-art 0.5B zero-shot text-to-speech model for expressive dialogue.

ChatTTS

2noise

Conversational text-to-speech for dialogue scenarios, in English and Chinese.

claude-context

zilliztech

A semantic code-search MCP with hybrid BM25 and vector retrieval.

claude-context-optimizer

egorfedorov

Tracks token usage, blocks redundant reads, and supports .contextignore and budget alerts.

claude-cookbooks

anthropics

Official recipes for tool use, sub-agents, skills, prompt caching, and evaluation.

claude-mem

thedotmack

Persistent cross-session memory that AI-compresses sessions to SQLite and Chroma.

claude-rolling-context

NodeNestor

A proxy plugin that rolls context compression past 100K tokens.

claude-skills

alirezarezvani

Bundles 345+ skills and plugins plus 579 CLI tools for Claude Code and 12 other agents.

claude-token-optimizer

nadimtuhin

Restructures CLAUDE.md and docs for roughly 90% token savings with a CLI audit and compress.

claw-lens

msfirebird

An open-source observability dashboard for OpenClaw daemons.

clawsec

prompt-security

A security skill suite for OpenClaw, Hermes, PicoClaw, and NanoClaw.

CogVLM2

THUDM

General VQA, document Q&A, and GUI understanding with high-resolution inputs

Conversational Speech Model (Sesame)

SesameAILabs

Multimodal model generating human-like conversational speech with autoregressive transformers.

cronalytics

8bit64k

A Hermes cron cost and observability plugin.

Dia 1.6B

Expressive text-to-speech with emotion and tone control plus nonverbal sounds.

Docling

docling-project

Document conversion with layout analysis and table structure recognition

DSPy

stanfordnlp

A framework for programming, not prompting, language models.

E2B

e2b-dev

Open-source secure cloud sandboxes for running AI-generated code.

Exa MCP server

exa-labs

An MCP giving agents Exa web and code search plus research.

Firecrawl

firecrawl

Turn web scrape, search, and crawl into an LLM-ready data API.

Giskard

Giskard-AI

Detect performance, bias, and security issues in AI applications.

GPT Researcher

assafelovic

Autonomous online research agent that scrapes, summarizes, and exports findings to PDF.

graph-memory

adoresever

A knowledge-graph context engine offering roughly 75% context compression and cross-session reuse.

Helicone

An open-source AI gateway and LLM observability proxy.

hermes-agent-desktop

desktop-hermes

A desktop client and UI for Hermes.

hermes-agent-self-evolution

NousResearch

Official evolutionary self-improvement that optimizes skills and prompts and opens PRs.

hermes-labyrinth

stainlu

A read-only observability black-box recorder that visualizes agent journeys.

hermes-lcm

stephenschoettler

Lossless Context Management for Hermes with a DAG context engine and lcm_grep and lcm_expand.

hermes-paperclip-adapter

NousResearch

Official adapter to run Hermes as a managed employee in a Paperclip company.

hermes-plugins

42-evey

A suite of 23 Hermes plugins for goals, inter-agent bridging, and cost control.

hermes-skill-factory

Romanescu11

A meta-skill that watches workflows and auto-creates reusable Hermes skills.

hermes-web-search-plus

robbyczgw-cla

Multi-provider web search and extract for Hermes.

hermes-workspace

outsourc-e

A native web workspace and UI for Hermes.

icarus-plugin

esaradev

A Hermes plugin for shared cross-instance memory and training-data extraction.

Inter-1

Social-intelligence API — 12 social signals from video, audio, and text.

laconic

GabrielBarberini

A small, minimal "be laconic" compression skill for roughly 90% prose-token reduction.

Langfuse

langfuse

Comprehensive LLM observability and debugging.

LangGraph

langchain-ai

Low-level orchestration for stateful agents.

LangMem

langchain-ai

Integrated long-term memory management for developers building on LangChain and LangGraph.

LLM Guard

protectai

An LLM security toolkit with input and output scanners.

Local Deep Research

LearningCircuit

Privacy-focused research assistant that runs iterative, cited analysis with local LLMs.

Mastra

mastra-ai

A TypeScript framework for AI apps and agents.

mcp-omnisearch

spences10

A single MCP unifying Tavily, Brave, Kagi, and Exa search.

Mem0

mem0ai

A universal memory layer that extracts salient facts and retrieves them at inference.

memory-lancedb-pro

CortexReach

A LanceDB long-term memory plugin with hybrid vector and BM25 search and cross-encoder rerank.

mission-control

builderz-labs

A self-hosted, framework-agnostic agent-fleet dashboard.

Molmo

allenai

Open-ended question answering and document reading, excelling at dense text like charts and forms

mPLUG-DocOwl2

X-PLUG

Multi-page document QA without OCR dependency via a high-resolution DocCompressor

NVIDIA Parakeet v2

High-quality English speech recognition with punctuation and word-level timestamping.

Ollama

Run open LLMs locally — no API key.

openclaw-a2a-gateway

win4r

An OpenClaw plugin implementing A2A v0.3.0 as a bidirectional gateway.

openclaw-channel-dingtalk

soimy

A DingTalk enterprise channel plugin for OpenClaw.

openclaw-composio-plugin

ComposioHQ

An OpenClaw plugin adding 1000+ third-party integrations.

opengoat

marian2js

Build organizations of OpenClaw agents coordinating across coding tools.

OpenLLMetry

traceloop

Observability for LLM applications using OpenTelemetry.

PaddleOCR

PaddlePaddle

Comprehensive OCR pipeline for multilingual tasks requiring high accuracy

Parler-TTS

Lightweight text-to-speech that generates natural speech in a given speaker style.

Phi-4 Multimodal

Multimodal document understanding with integrated speech and vision in a compact model

pi-coding-agent

dnouri

An Emacs frontend for the Pi coding agent.

Pipecat

pipecat-ai

Open-source Python framework for building real-time voice and multimodal conversational agents.

Playwright

microsoft

End-to-end browser automation with full control across Chromium, Firefox, and WebKit.

powermem

oceanbase

An AI memory plugin built on OceanBase for smarter agents.

Puppeteer

puppeteer

Lightweight Chrome automation with JavaScript and a rich developer ecosystem.

Qwen-2.5-Omni

Vision-language-audio model with speech input and output plus document understanding.

Qwen2.5-VL

QwenLM

Vision-language model for complex document QA, structured data extraction, and long-context processing

screenshot-to-code

abi

Turn screenshots into clean HTML, Tailwind, React, or Vue code for fast UI prototyping.

Smolagents

huggingface

A HuggingFace ~1k-LOC library for code-writing agents.

Speaker Diarization 3.1

Identify and segment speakers in audio, outputting speaker diarization annotations.

Stagehand

browserbase

Declarative automation of browser workflows via LLMs, described in plain language.

SuperClaude Framework

SuperClaude-Org

Adds 30 commands, 20 agents, 7 modes, and 8 MCP integrations to Claude Code.

Superpowers

obra

Agentic skills framework with a dev methodology for coding agents.

Tavily MCP

tavily-ai

A production MCP for real-time search, extract, map, and crawl.

tokenjuice

vincentkoc

An output-compaction and token-reduction CLI that runs standalone or as a framework extension.

Ultravox

Multimodal model for real-time voice interaction, consuming both speech and text inputs.

Voice Lab

saharmor

Framework for testing and evaluating voice agents across models, prompts, and personas.

wayid-nginx

LineageLabs

Verify the WayID of inbound AI agents at the NGINX edge, before traffic reaches your app.

wayid-verify

LineageLabs

Verify the WayID of a calling AI agent at your gateway, reverse proxy, or API backend.

wechat-openclaw-channel

HenryXiaoYang

A WeChat channel plugin with QClaw and WorkBuddy dual mode.

Whisper

openai

General-purpose speech recognition trained on a large dataset of diverse audio.

yantrikdb-hermes-plugin

yantrikos

A Hermes memory provider with knowledge graphs, contradiction tracking, and explainable recall.