@simonw

–

01 August 2025 6h42

Not as good, but it's also not useless which is saying a lot for a local coding model

Claude iOS File Attachment DataCloneError Bug Report

By

–

01 August 2025 5h22

I'm or getting 500 errors any more! I am quite reliably getting "Error: DataCloneError: The object can not be cloned." when I try to use file attachments (with mobile Safari embedded in the Claude iOS app) so something's not right about how it all works – demo:

Reverse Engineering System Prompts: 45 Posts Guide

By

–

01 August 2025 2h20

Reverse engineering system prompts is one of the best ways I know of to level up at prompt engineering – here's my collection of 45 posts on the topic dating all the way back to the Microsoft Bing prompt leak in February 2023

Claude Artifacts Now Enable AI-Powered Apps with API Access

By

–

01 August 2025 1h50

The second is the ability for artifacts to build "AI-powered apps" to which users can upload PDFs and images

What that means: Artifacts can call the real Claude API now via a monkey-patched fetch() that gets redirected through a proxy adding an API keyhttps://t.co/V5VJeR6SaQ
— Simon Willison (@simonw) 31 juillet 2025

The second is the ability for artifacts to build "AI-powered apps" to which users can upload PDFs and images What that means: Artifacts can call the real Claude API now via a monkey-patched fetch() that gets redirected through a proxy adding an API key

Claude Mobile Adds Native Calendar and Messaging Tools

By

–

01 August 2025 1h48

The first feature is the ability for Claude on mobile to trigger a human-in-the-loop native app dialog for adding a calendar event or sending a message or email – that's implemented using two new tools, event_create_v0 and message_compose_v0 https://t.co/qhksbvwxQj
— Simon Willison (@simonw) 31 juillet 2025

The first feature is the ability for Claude on mobile to trigger a human-in-the-loop native app dialog for adding a calendar event or sending a message or email – that's implemented using two new tools, event_create_v0 and message_compose_v0

Reverse Engineering Claude’s New Features Without Documentation

By

–

01 August 2025 1h47

Anthropic shipped two new features for Claude over the past couple of days, but as far as I can tell they didn't provide any documentation at all beyond a tweet about each one … so I reverse engineered the system prompt to figure out how they work

MCP servers security risks: protecting user data from theft

By

–

31 July 2025 23h04

Right now users are being encouraged to mix and match MCP servers in a way that makes it trivial for people to steal their data – we are unfairly outsourcing decisions on how to stay secure to people who haven't got a fighting chance of staying safe

Flawed AI Security Mitigations Encourage Inherently Insecure Systems

By

–

31 July 2025 22h57

When it comes to security I think flawed mitigations are actually harmful, because they encourage developers to build systems that are inherently insecure and can never be made safe for their users There are a LOT of things being built right now that should not be built

Prompt Injection Vulnerabilities Harder to Fix Than SQL Injection

By

–

31 July 2025 22h54

If I report a SQL injection vulnerability to a product they can reliably fix it and close the security hole If I report a prompt injection vulnerability they can't

SQL Security vs LLM Prompt Injection Vulnerabilities

By

–

31 July 2025 22h53

Re-read my post. We CAN achieve 100% security with SQL provided we don't make mistakes… and if we DO make mistakes we can fix them to get to 100% We can't do that with LLM prompt concatenation – we don't have a way to fix those mistakes! That's the difference here