prompt injection in local agent setups

[phasenode]

This keeps coming up in conversation so I figured I'd start a thread.

The more I work with local setups, the more it seems like small local stacks decides everything downstream.

Wondering if this is just where the whole space is heading now.

[llmarch51]

That matches what I was seeing. I don't think your conclusion is far off.

[vectorshed43]

That matches what I was seeing. I don't think your conclusion is far off.

Can you fill in a bit more context? Is this local-only or are you routing anything externally?

[sanescope]

A couple questions before I guess too hard. What model, what runtime, and roughly what context size?

[orbitbench6]

Did a small comparison this afternoon. The issue went away once I changed one setting and restarted the stack cleanly. The main thing that moved was latency, not quality.

[kernelhouse70]

Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[batchember]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there.

[promptspark58]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there.

Need a little more detail here. Are you testing this on M3 Max or something else?