Daniel Saito

Daniel Saito

Building Replay, AI agent cache audit
#102409620 followers0 following
🎉 6 day streak
>10,000All time
17 KP

About

Founder of Red Robot K.K., a software studio in Tokyo. I build Replay Doctor, a local command line tool that reads the transcripts coding agents already keep on disk and names the turn a prompt cache broke, the cause, and the tokens re-billed at write prices. It runs locally, sends nothing, and prints NOT MEASURED rather than estimating what it cannot measure. Every figure it publishes carries the population it was measured on and the date it was read, and when one turns out to be wrong the correction ships as a new dated file with the old reading still visible.

Badges

Tastemaker
Tastemaker
Gone streaking
Gone streaking
Gone streaking 5
Gone streaking 5

Maker History

  • Replay Doctor
    Replay DoctorFind the turn your prompt cache broke, and what it cost
    Sep 2026
    • 🎉
    Joined Product HuntSeptember 14th, 2026

Forums

Replay Doctorp/replay-ai-agent-cache-audit•

4h ago

Built to measure GPT-6 Astra. Not measured yet.

I will say the awkward part first. Replay can read what GPT-6 Astra writes, and it knows Astra's published pricing, but I have never measured a real Astra session. So I do not have an Astra number, and I am not going to produce one.

What exists in the shipped build: the reader for the rollout format Astra writes, and the pricing rules that go with it. The path is there and the code runs.

Replay Doctorp/replay-ai-agent-cache-audit•

4h ago

What is your re-billed percentage? Mine is 4.18%

On 123 sessions read 2026-09-17, 4.18% of my spend was tokens I paid for twice. That covers 105.2M tokens and $620.47 at list prices.

That is what a broken prompt cache does. The provider re-bills the whole conversation at write prices. Nothing errors. Nothing warns you. The bill just looks like you used the thing more.

Replay Doctorp/replay-ai-agent-cache-audit•

4h ago

Running GPT-6 Astra? Can REPLAY scan your corpus and get the statistical data?

I built Replay to find where prompt caches break and what that costs you. It can already read Astra's rollout files. The problem is I don't have any Astra traffic to point it at.

Here's the honest situation. Replay reads the same rollout format Astra writes, and it knows Astra's pricing and cache rules. But I've never actually measured an Astra session, because I can't afford to generate the traffic. When there's no data, Replay says NOT MEASURED instead of guessing. The site says the same thing.

View more