A korrentour readingWhat is a korrent?
Any LLM system that uses compaction needs to preserve the pre-compacted text and make it available via agent tool calls.
Drawn from what Simon Willison said
SubjectLLMs
What Simon Willison actually said
Word for word, with the source under each one. They did not write this page.
Co-creator of Django and creator of Datasette
I think any LLM system that uses compaction needs to both preserve the pre-compacted text and make that text available via agent tool calls, to protect against this kind of problem.
Where this was said
↗Generating running routes with GPT-6 Astra and ChatGPT Worksimonwillison.net
Added to korrents 12 Sept 2026 · How quotes work · Something wrong? Tell us
Do you hold this korrent?Do you also believe this?
Related korrents
Our reading — they may agree, disagree or merely touch the same thing. Closest first: a shared subject counts for most, then how near the wording is.