Merge branch 'main' of https://github.com/zachary62/miniLLMFlow

README.md

@@ -68,13 +68,14 @@ Current LLM frameworks are bloated... You only need 100 lines for LLM Framework!
   
 |  App Name     |  Difficulty    | Topics  | Human Design | Agent Code |
 | :-------------:  | :-------------: | :---------------------: |  :---: |  :---: |
+| [Build Cursor with Cursor](https://github.com/The-Pocket/Tutorial-Cursor) <br> <sup><sub>We’ll reach the singularity soon ...</sup></sub> | ★★★ <br> *Advanced*   | [Agent](https://the-pocket.github.io/PocketFlow/design_pattern/agent.html) | [Design Doc](https://github.com/The-Pocket/Tutorial-Cursor/blob/main/docs/design.md) | [Flow Code](https://github.com/The-Pocket/Tutorial-Cursor/blob/main/flow.py)
 | [Ask AI Paul Graham](https://github.com/The-Pocket/Tutorial-YC-Partner) <br> <sup><sub>Ask AI Paul Graham, in case you don't get in</sup></sub> | ★★☆ <br> *Medium*   | [RAG](https://the-pocket.github.io/PocketFlow/design_pattern/rag.html) <br> [Map Reduce](https://the-pocket.github.io/PocketFlow/design_pattern/mapreduce.html) <br> [TTS](https://the-pocket.github.io/PocketFlow/utility_function/text_to_speech.html) | [Design Doc](https://github.com/The-Pocket/Tutorial-AI-Paul-Graham/blob/main/docs/design.md) | [Flow Code](https://github.com/The-Pocket/Tutorial-AI-Paul-Graham/blob/main/flow.py)
 | [Youtube Summarizer](https://github.com/The-Pocket/Tutorial-Youtube-Made-Simple)  <br> <sup><sub> Explain YouTube Videos to you like you're 5 </sup></sub> | ★☆☆ <br> *Beginner*   | [Map Reduce](https://the-pocket.github.io/PocketFlow/design_pattern/mapreduce.html) |  [Design Doc](https://github.com/The-Pocket/Tutorial-Youtube-Made-Simple/blob/main/docs/design.md) | [Flow Code](https://github.com/The-Pocket/Tutorial-Youtube-Made-Simple/blob/main/flow.py)
 | [Cold Opener Generator](https://github.com/The-Pocket/Tutorial-Cold-Email-Personalization)  <br> <sup><sub> Instant icebreakers that turn cold leads hot </sup></sub> | ★☆☆ <br> *Beginner*   | [Map Reduce](https://the-pocket.github.io/PocketFlow/design_pattern/mapreduce.html) <br> [Web Search](https://the-pocket.github.io/PocketFlow/utility_function/websearch.html) |  [Design Doc](https://github.com/The-Pocket/Tutorial-Cold-Email-Personalization/blob/master/docs/design.md) | [Flow Code](https://github.com/The-Pocket/Tutorial-Cold-Email-Personalization/blob/master/flow.py)
 
 </div>
 
-- Want to learn **Agent Coding**? Check out [my YouTube](https://www.youtube.com/@ZacharyLLM?sub_confirmation=1)!
+- Want to learn **Agent Coding**? Check out [my YouTube](https://www.youtube.com/@ZacharyLLM?sub_confirmation=1)! Read this [Guide](https://the-pocket.github.io/PocketFlow/guide.html)!
   
 - Want to build your own LLM App? Start with [this template](https://github.com/The-Pocket/PocketFlow-Template-Python)!
 

docs/core_abstraction/communication.md

@@ -7,23 +7,23 @@ nav_order: 3
 
 # Communication
 
-Nodes and Flows **communicate** in two ways:
+Nodes and Flows **communicate** in 2 ways:
 
-1. **Shared Store (recommended)** 
+1. **Shared Store (for almost all the cases)** 
 
-   - A global data structure (often an in-mem dict) that all nodes can read and write by `prep()` and `post()`.  
+   - A global data structure (often an in-mem dict) that all nodes can read ( `prep()`) and write (`post()`).  
    - Great for data results, large content, or anything multiple nodes need.
    - You shall design the data structure and populate it ahead.
      
+   - > **Separation of Concerns:** Use `Shared Store` for almost all cases to separate *Data Schema* from *Compute Logic*!  This approach is both flexible and easy to manage, resulting in more maintainable code. `Params` is more a syntax sugar for [Batch](./batch.md).
+     {: .best-practice }
+
 2. **Params (only for [Batch](./batch.md))** 
    - Each node has a local, ephemeral `params` dict passed in by the **parent Flow**, used as an identifier for tasks. Parameter keys and values shall be **immutable**.
    - Good for identifiers like filenames or numeric IDs, in Batch mode.
 
 If you know memory management, think of the **Shared Store** like a **heap** (shared by all function calls), and **Params** like a **stack** (assigned by the caller).
 
-> Use `Shared Store` for almost all cases. It's flexible and easy to manage. It separates *Data Schema* from *Compute Logic*, making the code easier to maintain. `Params` is more a syntax sugar for [Batch](./batch.md).
-{: .best-practice }
-
 ---
 
 ## 1. Shared Store

docs/guide.md

@@ -26,7 +26,7 @@ Agentic Coding should be a collaboration between Human System Design and Agent I
     - suitable for routine tasks that require common sense (e.g., filling out forms, replying to emails).
     - suitable for creative tasks where all inputs are provided (e.g., building slides, writing SQL).
     - **NOT** suitable for tasks that are highly ambiguous and require complex info (e.g., building a startup).
-    - > **If a human can’t solve it, an LLM can’t automate it!** Before building an LLM system, thoroughly understand the problem by manually solving example inputs to develop intuition.  
+    - > **If Humans can’t specify it, AI Agents can’t automate it!** Before building an LLM system, thoroughly understand the problem by manually solving example inputs to develop intuition.  
       {: .best-practice }
 
 
@@ -68,11 +68,12 @@ Agentic Coding should be a collaboration between Human System Design and Agent I
 
 
 5. **Implementation**: Implement the initial nodes and flows based on the design.
+   - 🎉 If you’ve reached this step, humans have finished the design. Now *Agentic Coding* begins!
    - **“Keep it simple, stupid!”** Avoid complex features and full-scale type checking.
    - **FAIL FAST**! Avoid `try` logic so you can quickly identify any weak points in the system.
    - Add logging throughout the code to facilitate debugging.
 
-6. **Optimization**:
+7. **Optimization**:
    - **Use Intuition**: For a quick initial evaluation, human intuition is often a good start.
    - **Redesign Flow (Back to Step 3)**: Consider breaking down tasks further, introducing agentic decisions, or better managing input contexts.
    - If your flow design is already solid, move on to micro-optimizations:
@@ -84,7 +85,7 @@ Agentic Coding should be a collaboration between Human System Design and Agent I
      > <div align="center"><img src="https://github.com/the-pocket/PocketFlow/raw/main/assets/success.png?raw=true" width="400"/></div>
      {: .best-practice }
 
-7. **Reliability**  
+8. **Reliability**  
    - **Node Retries**: Add checks in the node `exec` to ensure outputs meet requirements, and consider increasing `max_retries` and `wait` times.
    - **Logging and Visualization**: Maintain logs of all attempts and visualize node results for easier debugging.
    - **Self-Evaluation**: Add a separate node (powered by an LLM) to review outputs when results are uncertain.