Automação
chat-on-steroids
Cross-platform local MCP capabilities for ChatGPT with Chrome integration, Goal, Compact & Resume, and durable multi-agent workflows.
★ 4,2 mil estrelasMIT
Ver o guia →agent-toolkit é um projeto de automação de código aberto de softaworks (softaworks/agent-toolkit no GitHub). Tem 2.520 estrelas no GitHub e 229 forks, é publicado sob a licença MIT e foi atualizado pela última vez no GitHub em 5 de março de 2026.
A curated collection of skills for AI coding agents. Skills are packaged instructions and scripts that extend agent capabilities across development, document…
Link the services you already use through built-in integrations or generic webhooks and API calls.
Drag triggers and actions onto a canvas: when this happens, do that.
Branch on conditions, loop over items, or pipe steps through language models.
Turn the workflow on and watch executions in the run history.
Designed to run unattended: triggers, schedules, and webhooks.
Export, diff, and back up workflows like code.
The automation is free; you only pay for your own server.
Cron, webhooks, and events keep flows running 24/7.
Call LLMs mid-workflow for decisions and extraction.
Credentials and execution data stay on your infrastructure.
Download the source and move into the project folder.
git clone https://github.com/softaworks/agent-toolkit.git
cd agent-toolkitEvery project documents its own setup; start with install requirements.
cat README.md | head -60Follow the project's launch command — Docker, pip, npm, or a binary release.
# see README for the exact launch commandAlso on this shelf: AdStrike — modular AD red-team framework with AI help for authorized pentesting. The Automation shelf also holds Well — No more Sundays on Finance. Readers working through the Automation shelf also stop at Everfern, a free AI agent that drives your computer like you would.
It is an open-source project at softaworks/agent-toolkit with 229 forks. Check the repo's contributors and recent commits to see how active maintenance is.
MIT licensed — use it commercially, no strings attached.
Docker and a small VPS or home server cover most of these.
Same idea, but self-hosted with no per-task pricing.
Most flows are visual; light expressions help for advanced logic.
Cross-platform local MCP capabilities for ChatGPT with Chrome integration, Goal, Compact & Resume, and durable multi-agent workflows.
A GitHub action to create a pull request for changes to your repository in the actions workspace
Event Driven Orchestration & Scheduling Platform for Mission Critical Applications
Automates everything around coding: plugins, agents, and skills for AI IDEs.
Skills giving AI agents live web data via search and extraction.
Drive real phones with AI agents — Android via ADB, iPhone via WebDriverAgent.
Os outros projetos de Automação com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se agent-toolkit não for o ideal para você.
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+…
Langflow — the visual builder for AI agents and workflows: drag-and-drop flows over LangChain components, deployed as APIs or MCP tools, MIT-licensed
A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows
Flowise — build AI agents visually: an open-source drag-and-drop UI over LangChain and LlamaIndex, with chatflows, agentflows and a marketplace of templates
Create agents that monitor and act on your behalf. Your agents are standing by!
Projetos de Automação próximos no catálogo, os mais próximos em estrelas de agent-toolkit: continue explorando dos dois lados dele.
Amazing Alfred Workflows
CodeMachine is an open-source tool that orchestrates AI coding agents into repeatable, long-running workflows. ⚡️
Customize your Zotero workflow.
Free hands-on course on using AI dev tools without losing engineering discipline.
A GitHub action to create a pull request for changes to your repository in the actions workspace
Full computer-use for AI agents on macOS — self-learning workflows, no screenshots.
Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.