awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
lukilabs avatar

lukilabs/craft-agents-oss

0
View on GitHub↗
2,888 stars·409 forks·TypeScript·apache-2.0·17 views

Craft Agents Oss

Craft Agents is an open-source desktop application that serves as a unified hub for managing multiple AI agents and their providers from a single interface. It connects to services including Anthropic, Google AI Studio, ChatGPT Plus, and GitHub Copilot, while also supporting any OpenAI-compatible endpoint, with the ability to set per-workspace provider defaults for each session.

The application operates on a client-server architecture that decouples the agent runtime into a remote server process and a thin desktop client, enabling sessions to persist across machines. It includes a WebSocket-based CLI interface for terminal-based agent interaction suitable for scripting and CI/CD pipeline integration, along with an event-driven automation engine that triggers agent workflows based on label changes, schedules, and tool-use events through a configurable rule system.

Craft Agents provides granular permission control that intercepts agent actions at runtime to enforce read-only, ask-to-edit, or auto-approve modes with dynamic mode cycling during conversations. The system integrates with external data sources through a protocol adapter that translates natural language requests into structured calls to Model Context Protocol servers, REST APIs, and local data sources, while also supporting natural language API discovery that automatically configures credentials for external services. Sessions are organized through a visual inbox with status workflows, flagging, and full history persistence to disk for continuity across restarts.

Features

  • Desktop AI Agents - Opens a native desktop application that manages connections to multiple AI agents and APIs through a graphical interface.
  • Agent Server Hosting - Operates the agent as a headless server on a remote machine for persistent, long-running sessions via a thin client.
  • Agent Connection Management - Connects to multiple AI agent services from a single desktop environment to manage all interactions.
  • Multi-Provider Orchestrators - Routes messages between a local desktop client and multiple remote AI agent endpoints through a unified connection manager.
  • Provider Switching - Switches between multiple AI providers and sets per-workspace defaults for each session.
  • Session State Persistence - Maintains full conversation history and state to disk for session continuity across restarts and machine switches.
  • MCP Protocol Adapters - Translates natural language requests into structured calls to Model Context Protocol servers, REST APIs, and local data sources.
  • Headless Agent Servers - Decouples the agent runtime into a remote server process and a thin desktop client for persistent sessions across machines.
  • Agent Permission Modes - Intercepts agent actions at runtime to enforce read-only, ask-to-edit, or auto-approve modes with dynamic cycling.
  • Visual Inboxes - Provides a visual inbox with status workflows, flagging, and full history persistence for organizing agent conversations.
  • Conversation Organizers - Organizes conversations with status workflows, flagging, AI-generated titles, and full history persistence to disk.
  • External Data Integrations - Pulls data from MCP servers, REST APIs, local files, and databases through natural language requests or configuration.
  • Workspace-Specific Provider Selections - Picks from multiple AI providers to assign a specific backend per workspace.
  • Event-Driven Automation Engines - Triggers agent sessions and workflows by monitoring label changes, schedules, and tool-use events.
  • Workflow Event Triggers - Triggers automated agent sessions using natural language prompts when events like label changes or schedules occur.
  • WebSocket-Based CLI Exposure - Connects a terminal client to a running agent server over WebSocket for scripting and CI/CD pipeline integration.
  • Automatic Credential Configurations - Describes an external service in natural language so the agent discovers and configures the API automatically.
  • Desktop Inbox Managers - Manages multiple concurrent agent conversations with an inbox, status workflow, flagging, and full history saved to disk.

Star history

Star history chart for lukilabs/craft-agents-ossStar history chart for lukilabs/craft-agents-oss

How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI

Projects sharing features with Craft Agents Oss

These projects share indexed features with Craft Agents Oss. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
  • doriandarko/maestroDoriandarko avatar

    Doriandarko/maestro

    4,320View on GitHub↗

    Maestro is an autonomous task workflow engine that decomposes high-level goals into hierarchical sub-tasks and orchestrates their execution using multiple language model agents. It provides a unified interface for routing requests across different LLM providers, including proprietary models like Anthropic, OpenAI, and Gemini, as well as local models, enabling flexible provider selection and switching through a single entry point. The system distinguishes itself through its ability to generate complete software project structures directly on the host machine, creating directories and source fi

    Python
    View on GitHub↗4,320
  • muesli/beehivemuesli avatar

    muesli/beehive

    6,476View on GitHub↗

    Beehive is a self-hosted integration platform and event automation engine. It functions as a plugin-based workflow orchestrator that links independent service plugins to automate data movement between external APIs and hardware. The system acts as an IoT hardware controller for managing network-enabled sockets, smart lighting, and serial port devices via TTY or COM. It also serves as an HTTP webhook gateway, receiving incoming requests to trigger internal workflows and external notifications. Capability areas include event-driven workflow automation with filter-based routing and template-bas

    Go
    View on GitHub↗6,476
  • kestra-io/kestrakestra-io avatar

    kestra-io/kestra

    27,073View on GitHub↗

    Kestra is a declarative workflow orchestrator designed to manage complex task dependencies and automated processes through versioned configuration files. It functions as a distributed platform that decouples task scheduling from execution by offloading computational workloads to a fleet of worker nodes. The system uses a reactive, event-driven engine to initiate workflows automatically in response to external signals, webhooks, schedules, or file system changes. The platform distinguishes itself through a modular plugin architecture that allows for the integration of custom tasks and external

    Javaautomationdata-orchestrationdevops
    View on GitHub↗27,073
  • getpaseo/paseogetpaseo avatar

    getpaseo/paseo

    9,118View on GitHub↗

    Paseo is an LLM coding agent orchestrator and multi-agent workflow manager designed to coordinate multiple AI agents across isolated git worktrees. It provides a unified control interface for managing these agents and their associated environments to execute complex programming tasks. The system distinguishes itself through a remote agent daemon that enables secure access to local coding agents via encrypted relays. It employs a git worktree environment manager to isolate parallel tasks into dedicated directories and branch-based server URLs, preventing file collisions and network port confli

    TypeScriptadeagentsclaude-code
    View on GitHub↗9,118
Compare all 30 related projects→

Frequently asked questions

What does lukilabs/craft-agents-oss do?

Craft Agents is an open-source desktop application that serves as a unified hub for managing multiple AI agents and their providers from a single interface. It connects to services including Anthropic, Google AI Studio, ChatGPT Plus, and GitHub Copilot, while also supporting any OpenAI-compatible endpoint, with the ability to set per-workspace provider defaults for each session.

What are the main features of lukilabs/craft-agents-oss?

The main features of lukilabs/craft-agents-oss are: Desktop AI Agents, Agent Server Hosting, Agent Connection Management, Multi-Provider Orchestrators, Provider Switching, Session State Persistence, MCP Protocol Adapters, Headless Agent Servers.

Which projects share features with lukilabs/craft-agents-oss?

Projects with overlapping indexed features include: doriandarko/maestro — Maestro is an autonomous task workflow engine that decomposes high-level goals into hierarchical sub-tasks and… muesli/beehive — Beehive is a self-hosted integration platform and event automation engine. It functions as a plugin-based workflow… kestra-io/kestra — Kestra is a declarative workflow orchestrator designed to manage complex task dependencies and automated processes… getpaseo/paseo — Paseo is an LLM coding agent orchestrator and multi-agent workflow manager designed to coordinate multiple AI agents… argoproj/argo — Argo is a cloud native CI/CD platform and Kubernetes workflow engine. It functions as a container pipeline… inngest/inngest — Inngest is a durable execution framework and event-driven automation engine designed to orchestrate background…