AI Chat

Always benchmark before and after

No context or behavior change ships without an ai_evals A/B on the affected mode. Add or adjust cases for exactly what you changed, see the ai-evals skill for authoring and the full run reference.

Run the affected mode before your change and after, same model(s), same cases.

Where guidance goes

What the model should know about Windmill itself (flow shapes, groups, data tables, app access, secrets…) belongs in system_prompts/base/ or system_prompts/languages/, which also feed the CLI's skills, not in a */core.ts prompt builder, where only this chat would see it. The builders add tool plumbing and runtime values only. Scope a sentence to one consumer with <!-- chat-only --> / <!-- cli-only -->, keep chat tool names out of the shared files, and rerun python system_prompts/generate.py. system_prompts/README.md has the mechanics.

Measure the window first, and cumulative second

Optimize finalContextTokens (window occupancy, what drives overflow and compaction), then cumulative prompt tokens.

Context discipline

The dominant fixed cost is per-iteration overhead: the system prompt plus every tool schema is re-sent on every loop iteration. So:

• Every tool and every parameter is a permanent tax. Justify each one and measure it; an extra "locate" round-trip can cost more than the reads it saves. Strip dead params rather than leaving them in the schema.
• Tool results return the minimum. Never echo content the model already has. The canonical mistake: a write tool that returns the whole edited artifact right after the model authored it, return { success, message } instead. When you touch a shared write helper (e.g. finishAppDraftWrite in global/core.ts), re-check this invariant for all the write tools routing through it, the echo has regressed before via a shared refactor.

Prompts and tool descriptions are part of the surface