Skip to content

Instantly share code, notes, and snippets.

View charl-kruger's full-sized avatar

Charl Kruger charl-kruger

  • Investec
  • South Africa
View GitHub Profile
there is some strange issue where I post a message from whatsapp, then I post another. once I post the seconds, only then does it start │
│ analysing the first message.\ │
│ \ │
│ Deeply analyse the flow to understand what could be going wrong.\ │
│ \ │
│ this also happens for images. i upload an image, then a second. only once I upload the second does the first one show the processing etc
User approved Claude's plan:
⎿ Simple Hackathon Fix for Concurrent Processing
Problem
Users can upload multiple images before processing completes, causing confusion with multiple AI responses and choice prompts.
Solution (Quick & Simple)
Add processing state guards at key points:
> ok while we are still busy prosessing an image, we shouldnt take on new image processing untill we are done. e.g. when I upload a image, i │
│ dont get a fast response the I upload another... then I get two responses so the ai is getting confused, what should we do? Also remember │
│ this is a hackathon so we dont want to be too fancy. if need be, just tell the user to wait..
Ready to code? │
│ │
│ Here is Claude's plan: │
│ ╭────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮ │
│ │ FASHN WhatsApp Agent with Cloudflare Agents & Scheduled Tasks │ │
│ │ │ │
│ │ Architecture Overview │
Ready to code? │
│ │
│ Here is Claude's plan: │
│ ╭────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮ │
│ │ FASHN WhatsApp Agent with Cloudflare Agents & Scheduled Tasks │ │
│ │ │ │
│ │ Architecture Overview │
ok now that you fully understand how cloudflare agnets work, we want to update our agent to recieve a image via whatsapp (have a look at │
│ '/Users/charl.kruger/gpt5/fashn-hackathon/packages/fashn-whatsapp-agent' as a refrence of how to do the whatsapp webhook and recieve │
│ images), then we can ask the user what they want to do with it via ai.\ │
│ \ │
│ the tree use cases are:\ │
│ \ │
│ store the image as a profile pic, we will store this in cloudflare r2 and use this image if the user wants to do a model swap using the │
deeply analylse cloudflare agents https://developers.cloudflare.com/agents/ and pay special attention to Schedule tasks as we will defs be
nneding this.\
\
Once you fully understand how cloudflare agents work, analyse out new app here
'/Users/charl.kruger/gpt5/fashn-hackathon/packages/agents-fashn'
@charl-kruger
charl-kruger / yaml
Last active July 10, 2025 07:09
v0 API Swagger
openapi: 3.0.3
info:
title: V0 API
version: 1.1.0
description: |
Swagger spec generated from the v0 JavaScript client
**with concrete TypeScript definitions mapped to JSON-schema**.
servers:
- url: https://api.v0.dev/v1
# Build Agents on Cloudflare
URL: https://developers.cloudflare.com/agents/
import {
CardGrid,
Description,
Feature,
LinkButton,
LinkTitleCard,
{
"name": "AI product Images",
"nodes": [
{
"parameters": {},
"type": "n8n-nodes-base.manualTrigger",
"typeVersion": 1,
"position": [
0,
0