I saw the Notion MCP Challenge at the start of march and wanted to participate. But due to less time and some other stuff, i forgot about it, and when DEV mailed me about their events, i suddenly remembered i had to do it. In these last 10 days, i have been working [along with my AI companions] to try to make something crazzzy good. We finally did it and......
But before you need to know smth about me. I am Rudra, a 15-year-old school-going developer from India. [This also explains the time constraints].
So before submission i decided to again look at the rules. And guess what i saw..... Its for 18+ 😭😭😭😭😭😭😭😭😭😭😭. i did all this or nothing. But this isnt the end. So guys pls see this repo and give some feedback.
Also why are there age constraints on coding. i mean there have to be some, but them pls do smth or students too. We wanna win and earn smth too.
Pls see this repo
Rudra070311
/
NotionOS_X
Autonomous multi-agent AI operating system framework with DAG planning, persistent 3-tier memory, critic ensemble evaluation, self-improvement loops, and human-in-the-loop governance.
NotionOS X - Production-Grade AI Operating System
An advanced, fully-functional multi-agent AI operating system that simulates autonomous execution with persistent intelligence, memory systems, and human-in-the-loop governance.
Features
1. Task Graph Engine (DAG)
- Converts natural language tasks into Directed Acyclic Graphs
- Supports sequential and parallel execution stages
- Validates for cycles and structural integrity
- Optimizes for efficient execution order
2. Multi-Agent System
- Planner Agent: Generates execution plans and DAGs
- Executor Agent: Executes tasks and manages tool calls
- Researcher Agent: Gathers external knowledge and context
-
Critic Agents (3 variants):
- Quality Critic: Evaluates correctness and completeness
- Creativity Critic: Assesses novelty and innovation
- Practicality Critic: Measures applicability and feasibility
3. Execution Engine
- Async execution with configurable parallelism
- Respects DAG dependencies automatically
- Handles failures gracefully
- Tracks execution metrics
4. Persistent Memory System
- Episodic Memory: Records past task executions
- Feedback Memory: Stores critic scores and improvements
- Strategy Memory: Maintains learned patterns…
Will give an update on using it soon [Everuthing is in the repo already]
Top comments (0)