Adaptive Slider CAPTCHA Solver
Budget: $30 – $250 USD
I need code written will consistently solve the visual slider challenges served from geo.captcha-delivery.com (PerimeterX). The solution has to adapt to any script or site that embeds this CAPTCHA, not just a single hard-coded implementation, and it must stay resilient when PerimeterX updates its checks or tries to blacklist automated patterns.
My main target is the Visual slider CAPTCHA; no text or image puzzles are in scope. I’m still weighing how “human” the interaction needs to appear, so part of the job is to walk me through trade-offs between full behavioural emulation (variable mouse speed, pauses, noise) and a leaner, faster approach that still slips past PerimeterX’s bot profiler. I’ll make the final call once I see the feasibility and performance impact.
Here’s what I expect back:
• Well-structured, heavily commented source code.
• A clear README explaining setup, required libraries, and how to plug the solver into any headless or GUI browser flow.
• A short demo script showing the solver beating at least ten consecutive slider challenges in real time.
• Notes on how to retrain or tweak the solution when PerimeterX rolls out a change.
If you lean on machine-learning libraries (OpenCV, TensorFlow, scikit-image) or algorithmic tricks (edge detection, template matching, motion simulation), mention your plan up front so I can provision the right environment.
Success is when I can drop the code into my pipeline, point it at a page protected by geo.captcha-delivery.com, and watch it pass the slider puzzle reliably without triggering blocks or rate-limit errors.
My main target is the Visual slider CAPTCHA; no text or image puzzles are in scope. I’m still weighing how “human” the interaction needs to appear, so part of the job is to walk me through trade-offs between full behavioural emulation (variable mouse speed, pauses, noise) and a leaner, faster approach that still slips past PerimeterX’s bot profiler. I’ll make the final call once I see the feasibility and performance impact.
Here’s what I expect back:
• Well-structured, heavily commented source code.
• A clear README explaining setup, required libraries, and how to plug the solver into any headless or GUI browser flow.
• A short demo script showing the solver beating at least ten consecutive slider challenges in real time.
• Notes on how to retrain or tweak the solution when PerimeterX rolls out a change.
If you lean on machine-learning libraries (OpenCV, TensorFlow, scikit-image) or algorithmic tricks (edge detection, template matching, motion simulation), mention your plan up front so I can provision the right environment.
Success is when I can drop the code into my pipeline, point it at a page protected by geo.captcha-delivery.com, and watch it pass the slider puzzle reliably without triggering blocks or rate-limit errors.