ИИ-агенты
unstructured
Turn messy documents into clean, structured data for language models.
★ 15,5 тыс. звёздApache-2.0
Смотреть руководство →mcp-shrimp-task-manager — проект с открытым исходным кодом из категории «ИИ-агенты» от cjo4m06 (cjo4m06/mcp-shrimp-task-manager на GitHub). У него 2 147 звёзд на GitHub и 254 форка, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 21 августа 2025.
Shrimp Task Manager: an MCP server that turns natural language into structured, dependent dev tasks.
Decide which project you want an agent to plan and execute in structured steps rather than ad hoc prompts.
Clone the repository, install dependencies, and build the project to produce the server entry point.
Add the shrimp-task-manager entry to your MCP client's configuration with a data directory and template language, then start the client with that config.
Initialise project rules, plan your first task in natural language, and execute tasks one by one or in continuous mode.
Analyses requirements and breaks complex projects into atomic, testable subtasks automatically.
Manages relationships between tasks and tracks status in real time as work proceeds.
Tasks, progress, and history are stored and backed up, so agents keep context despite token limits or new sessions.
A systematic research mode explores technologies, and project rules keep coding standards consistent across tasks.
Different AI agents can be assigned to specific tasks, with reflection steps to review and improve completed work.
A React task viewer with drag-and-drop and search, plus a lightweight web GUI enabled through configuration.
Node.js 18 or newer is required.
git clone https://github.com/cjo4m06/mcp-shrimp-task-manager.git
cd mcp-shrimp-task-manager
npm install && npm run buildAdd the server to a .mcp.json in your project with your data directory, as shown in the README.
# .mcp.json: command "node", args ["/path/to/mcp-shrimp-task-manager/dist/index.js"], env DATA_DIR, TEMPLATES_USE=en, ENABLE_GUI=falseLaunch your client with the config and use the natural-language commands.
claude --mcp-config .mcp.json
# then ask: plan task: implement user authenticationIt was created by cjo4m06 and is maintained with community contributions on GitHub.
Yes, it is MIT-licensed and free; you use it through your existing MCP client and its model access.
Shrimp itself needs no API key; the AI client you connect it to (such as Claude Code) uses your own account or key.
A plain to-do list records tasks; Shrimp is an MCP server that actively plans, decomposes, orders by dependency, and reflects on tasks inside the agent's workflow.
Clone the repo, npm install, npm run build, add the server entry to your client's MCP config, and ask it to plan a task.
Turn messy documents into clean, structured data for language models.
Bright Data's MCP server: web search, scraping, structured data and browser automation for agents (75 tools)
Open skill that turns novels and stories into multi-episode video storyboards.
Bridge AI assistants to the Unity Editor via MCP: manage assets, scenes, scripts and tasks.
Google's MCP Toolbox for Databases - give agents safe, structured access to your data
Build production-ready agentic workflows from natural language — agents that operate browsers and computers.
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если mcp-shrimp-task-manager вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к mcp-shrimp-task-manager — продолжайте изучать полку по обе стороны от него.
A benchmark for tool-agent-user interaction in real-world domains.
One open platform to simulate, evaluate, guardrail, trace, and optimise LLM and agent apps, self-hostable.
Open Cowork: a desktop AI agent app with one-click Claude Code, MCP tools, skills, and VM sandboxing.
Open-source AI assistant framework with skills and an OpenClaw-style agent core.
Self-hosted AI vulnerability research: agents find, validate, deduplicate, and rank real security issues.
Uber's ADR: discover, observe, benchmark, and detect threats across enterprise AI agents, deployed at Uber.
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.