路 security
The Terminology Problem Causing Security Teams Real Risks
Jailbreaks target the model's safety training; prompt injection hijacks application trust boundaries. Conflating them leads to defenses that miss your actual threat surface.
Blog
Thoughts on cybersecurity, web development, and learning
- 路 security
What Makes Indirect Prompt Injections Attacks Succeed
Anatomy of an Indirect Prompt Injection
Sane Python Debugging in VSCode
A path towards a more reasonable dev experience
Mastering Stream Detection: Using MutationObserver to Track LLM Responses in Real-Time
A vanilla JavaScript solution for detecting when streaming responses from Large Language Models have completed
Dealing with Direct Upload Attachments in Tests
The process of directly uploading an attachment means that you are attaching a blob using the blob's `signed_id`. Read on to see how to access a blob's signed_id before the blob is attached to your ActiveRecord model instance in an RSpec suite.
