how-to-solve-bls-captcha-step-by-step

BLS CAPTCHA — CaptchaAI Example Pack

Full working example for solving BLS CAPTCHA using the CaptchaAI API.

Languages: Python, Node.js, PHP, Go, Java, C#, Ruby, Rust, Kotlin, Bash

Related article

This example accompanies the blog article: How to Solve BLS CAPTCHA Step by Step

Prerequisites

• A CaptchaAI account with API key (get one here)
• A CAPTCHA image file for testing (PNG, JPG, etc.)
• One or more of: Python 3.8+, Node.js 16+, PHP 8.0+, Go 1.21+, Java 11+, .NET 8+, Ruby 3.0+, Rust 1.70+, Kotlin 1.9+, or Bash with curl and jq

BLS CAPTCHA parameters

Parameter	Required	Description
method	Yes	Must be bls
body	Yes	Base64-encoded CAPTCHA image data
textinstructions	Yes	Text instructions describing what to select (e.g. "Select all images with a car")

Response

The API returns the selected cell indices in the request field (e.g. "1,3,5").

Setup

1. Clone the repository:

   git clone https://github.com/CaptchaAI/CaptchaAI-Examples.git
   cd CaptchaAI-Examples/articles/how-to-solve-bls-captcha-step-by-step

2. Copy the environment file and add your credentials:

   cp .env.example .env

3. Edit .env with your API key, image file path, and instructions.

4. Follow the language-specific setup below.

Python

cd python
pip install -r requirements.txt
python solve.py

Node.js

cd node
npm install
node solve.js

PHP

cd php
composer install
php solve.php

Go

cd go
go run solve.go

Java

cd java
javac Solve.java
java Solve

C#

cd csharp
dotnet run

Ruby

cd ruby
ruby solve.rb

Rust

cd rust
cargo run

Kotlin

cd kotlin
kotlinc solve.kt -include-runtime -d solve.jar
java -jar solve.jar

Bash

cd bash
chmod +x solve.sh
bash solve.sh

How it works

1. The script loads the CAPTCHA image from the file specified in CAPTCHA_IMAGE_FILE.
2. The image is base64-encoded and sent via POST to in.php with method=bls and textinstructions.
3. The API returns a task ID.
4. The script polls res.php until the solution is ready.
5. The answer contains the selected cell indices (e.g. "1,3,5").

Notes

• The image file must exist and be readable.
• Submission uses POST (not GET) because the base64-encoded image body is too large for a URL.
• The textinstructions parameter tells the solver what to look for in the image grid.