I think my plan of never letting an LLM access the open Internet wins yet again. Maybe GET requests only, no headers or parameters, ask me for every new domain, and only URLs comprising words/numbers no alphanumerics. Maybe. But it seems trivial to allow for a prompt injected attack vector either way
https://github.com/anthropics/claude-code/issues/53231
- April 25, 2026
I think my plan of never letting an LLM access the open Internet wins yet again. Maybe GET requests only, no headers or parameters, ask me for every new domain, and only URLs comprising words/numbers no alphanumerics. Maybe. But it seems trivial to allow for a prompt injected attack vector either way
Claude will/did set the UA to your email on a GET
Yeah but that was because it was gently asked to please not. I mean running a firewall/proxy that actually does it.