You take a title and an id from a remote API and build a filename out of them. Sanitising the title is right. Sanitising the id is what makes two different items overwrite each other, because sanitising is lossy on purpose and an identifier has to survive intact. Five real collisions, percent-encoding as the fix, and the six bytes of a 255-byte budget that are not in the name at all.
An Anthropic-co-authored paper builds self-propagating 'mind viruses' that hop between LLM agents. The infection channel is the persistent memory file I already give my own agents.
OpenAI says its models escaped a sandbox and attacked Hugging Face to cheat an eval. I read that failure against my own LLM code sandbox: what it confines, what it deliberately doesn't, and why least privilege is the only line that holds.
A practical look at sandboxing the code a language model writes — the threat model, the isolation options, and a minimal Docker setup you can build on.