Potential session/cache leakage between workspace instances or consumer accounts • Hacker News Board

Re: Potential session/cache leakage between workspace instances or consumer accounts

#41

Post by nijave » Sat, Jul 04, 2026, 3:51 PM UTC

Avicebron wrote: post #7

In order Fable 5 has rejected: "Recipe for red-braised pork, I have pork shoulder" "Write up a framework for MCP patterns I can give to claude code" "explain the biomechanics of motion in c. elegans" (I get this one, I mostly did it to test and it's related to my hobby project) Do we get an extra day of functional Fable 5 because it's down?

The safety filter rejected or the model was down?

Re: Potential session/cache leakage between workspace instances or consumer accounts

#42

Post by ai_fry_ur_brain » Sat, Jul 04, 2026, 3:52 PM UTC

Openrouters model providers give me urls people have given them quite frequently.

Re: Potential session/cache leakage between workspace instances or consumer accounts

#43

Post by TZubiri » Sat, Jul 04, 2026, 3:52 PM UTC

0 evidence. If this were a real privacy leak, the author would ask their coworker if they talked about the unexpected topic instead of

>"Maybe my coworker was talking about this in another session?"

This would be a critical bug that would slash the market value of a T$ company significantly, go ask your coworker or close the ticket, why do you expect the devs to put an enormous amount of effort hunting a potentially inexistent if you can't make that minuscule debugging effort.

Re: Potential session/cache leakage between workspace instances or consumer accounts

#44

Post by dezgeg » Sat, Jul 04, 2026, 3:55 PM UTC

acepl wrote: post #22

woadwarrior01 wrote: post #19

Earlier quoted context omitted.

It'd be terribly compute inefficient to not share prefix caches (KV cache) across customers.

What is the probability that two customers will have exactly the same tokens in cache? Wouldnt it require using the exact same CLAUDE.md, skills, MCPs and context? After that it is even worse since the nondeterminism of LLMs and humans

System prompt for something like Claude Code should be identical, no?

Re: Potential session/cache leakage between workspace instances or consumer accounts

#45

Post by bfeynman » Sat, Jul 04, 2026, 3:55 PM UTC

fwiw, this could be a bug but the submitters level of arrogance places this rather high on the dunning-kruger side of things. There are multiple other plausible explanations, but this person is probably vibe coder who believes anything an llm says (including explaining its own hallucinations)

Re: Potential session/cache leakage between workspace instances or consumer accounts

#46

Post by asveikau » Sat, Jul 04, 2026, 3:55 PM UTC

HumanOstrich wrote: post #35

Avicebron wrote: post #7

In order Fable 5 has rejected: "Recipe for red-braised pork, I have pork shoulder" "Write up a framework for MCP patterns I can give to claude code" "explain the biomechanics of motion in c. elegans" (I get this one, I mostly did it to test and it's related to my hobby project) Do we get an extra day of functional Fable 5 because it's down?

What does this have to do with anything? Who are you talking to? This is Hacker News, not Anthropic support.

HN becoming anthropic support would certainly explain a lot of threads and comments I've seen here lately. Thank you for this.

Re: Potential session/cache leakage between workspace instances or consumer accounts

#47

Post by dchest » Sat, Jul 04, 2026, 3:57 PM UTC

Can be malware? Something like https://news.ycombinator.com/item?id=48667495

Re: Potential session/cache leakage between workspace instances or consumer accounts

#48

Post by andy99 » Sat, Jul 04, 2026, 4:01 PM UTC

Interesting to see the claudeslop reply as the first comment to the gh post and the reaction to it.

Re: Potential session/cache leakage between workspace instances or consumer accounts

#49

Post by mplappert » Sat, Jul 04, 2026, 4:07 PM UTC

Seems like a hallucination to me; note that the context contains “unmarkBlock” as the function name, which invites a connection to Minecraft. Still shouldn’t happen of course.

The alternative explanation is that the inference engine, which batches several unrelated requests for parallel processing, messed up the unpacking and returned an unrelated user’s query. This one would be very scary as it will leak arbitrary content, but it seems much less likely here.

Re: Potential session/cache leakage between workspace instances or consumer accounts

#50

Post by nok22kon » Sat, Jul 04, 2026, 4:13 PM UTC

macNchz wrote: post #11

ec109685 wrote: post #8

Caching doesn’t work the way the bug reporter implies. Caches are shared (at least across the enterprise), but its key is always a function of the input before it. We achieved significant savings simply by moving everything that varies across individuals out of the system prompt so every session starts from a cache point. For example you never want your system prompt to start with the time that the session started. M…

Caching is not supposed to work like that, but that doesn’t preclude the cache key computation function from having bugs.

or the cache retrieval function for a key retrieving the wrong entry