Spark
Sign In
Prompt Chain

Generate and Refine Code with Self-Correction

LangGraph workflow demonstrating code generation concepts inspired by AlphaCodium, including documentation routing, inline unit tests, and orchestration using

Copy chain
Works with githubmistral
LangGraph
Maintainer?
91
Spark score
out of 100
Updated 3 months ago
Version 1.0.0
Models
mistral large
Add to Favorites

Why it matters

Automate code generation by iteratively constructing, testing, and improving code snippets. This asset leverages a self-correction loop to enhance code quality and correctness.

Outcomes

What it gets done

01

Generate initial code based on user prompts.

02

Perform inline unit tests to validate code execution.

03

Iteratively refine code based on test results and self-correction.

04

Route documentation to aid in code generation.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/lg-langgraphcodeassistantmistral | bash

Steps

Steps in the chain

01
Route user questions to documentation types

Show how to route user questions to different types of documentation

02
Perform inline unit tests

Perform inline unit tests to confirm imports and code execution work

03
Orchestrate with LangGraph

Use LangGraph to orchestrate the code generation and testing workflow

Overview

Code generation with self-correction

What it does

A LangGraph demonstration implementing AlphaCodium-inspired concepts for code generation, including documentation routing and inline unit testing with the Mistral Codestral model.

How it connects

Use this when you want to explore LangGraph orchestration patterns for code generation workflows that include documentation routing and inline unit testing capabilities.

Source README

This directory is retained purely for archival purposes and is no longer updated. Please see the newly consolidated LangChain documentation for the most current information and resources.

Code generation with self-correction

AlphaCodium presented an approach for code generation that uses control flow.

Main idea: construct an answer to a coding question iteratively..

AlphaCodium iteravely tests and improves an answer on public and AI-generated tests for a particular question.

We will implement some of these ideas from scratch using LangGraph:

  1. We show how to route user questions to different types of documentation
  2. We we will show how to perform inline unit tests to confirm imports and code execution work
  3. We will show how to use LangGraph to orchestrate this

LLM

We'll use the Mistral API and Codestral instruct model, which support tool use!

Tracing

Optionally, we'll use LangSmith for tracing.

Code Generation

Test with structured output.

State

Graph

Trace:

https://smith.langchain.com/public/53bcdaab-e3c5-4423-9908-c44595325c38/r

Trace:

https://smith.langchain.com/public/e749936d-7746-49de-b980-c41b17986e79/r

Trace:

https://smith.langchain.com/public/f5c19708-7592-4512-9f00-9696ab34a9eb/r

Trace w/ good example of self-correction:

https://smith.langchain.com/public/b54778a0-d267-4f09-bc28-71761201c522/r

Trace w/ good example of failure to correct:

https://smith.langchain.com/public/871ae736-2f77-44d4-b0da-a600d8f5377d/r

Step 1: Route user questions to documentation types

Show how to route user questions to different types of documentation

Step 2: Perform inline unit tests

Perform inline unit tests to confirm imports and code execution work

Step 3: Orchestrate with LangGraph

Use LangGraph to orchestrate the code generation and testing workflow

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.