Caching the outline cut reporting-agent deck generations by roughly two-thirds

manan_todi44 · reddit · 2026-07-21

After running a recurring reporting agent for months, the author found that the biggest cost saver was not prompt tuning but caching the stable outline in the middle of the pipeline.

What changed

The agent still reads weekly data, reasons about changes, and produces a status deck.
Instead of regenerating everything on every run, it now hashes a structured outline: headline plus three bullet points per section, with a hard length cap.
If the underlying data has not changed enough to alter that outline, the render step is skipped and the previous deck link is reused.

Result

Regenerations dropped by roughly two-thirds.
Deck quality stayed the same, because the outline was the part that mattered.
The key constraint was a small generation credit pool, around 50 renders per month on the author’s plan.

The general rule: the expensive part of an agent is often the piece you let it redo when nothing changed. Cache the stable artifact and only pay for the delta.

Original post →

Caching the outline cut reporting-agent deck generations by roughly two-thirds

More from coding & agent