BRbrudgers·Sep 3, 2025
Code page at github https://github.com/byme8/apparatus.exec
In Improving LLM token usage when debuggingbyme8
apparatus.exec
TypeScript2025–2025
byme8 / apparatus.exec
apparatus.exec
Discover this repository through its Hacker News conversations.
2Distinct people
3Mentions
0Explicit recommenders
16GitHub stars
What to know before exploring
What it is
A public description has not been verified.
Where it is used
No documented product use has been verified yet.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 3:56:07 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by 3-hour interval3 dated mentions
Observed mentions: 2025-09-01 13:15:06 UTC to 2025-09-03 19:01:33 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| Sep 1, 12:00 | 2025-09-01 12:00:00 UTC | 2025-09-01 15:00:00 UTC | 2 | 1 | 1 |
| Sep 1, 15:00 | 2025-09-01 15:00:00 UTC | 2025-09-01 18:00:00 UTC | 0 | 0 | 0 |
| Sep 1, 18:00 | 2025-09-01 18:00:00 UTC | 2025-09-01 21:00:00 UTC | 0 | 0 | 0 |
| Sep 1, 21:00 | 2025-09-01 21:00:00 UTC | 2025-09-02 00:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 00:00 | 2025-09-02 00:00:00 UTC | 2025-09-02 03:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 03:00 | 2025-09-02 03:00:00 UTC | 2025-09-02 06:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 06:00 | 2025-09-02 06:00:00 UTC | 2025-09-02 09:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 09:00 | 2025-09-02 09:00:00 UTC | 2025-09-02 12:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 12:00 | 2025-09-02 12:00:00 UTC | 2025-09-02 15:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 15:00 | 2025-09-02 15:00:00 UTC | 2025-09-02 18:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 18:00 | 2025-09-02 18:00:00 UTC | 2025-09-02 21:00:00 UTC | 0 | 0 | 0 |
| Sep 2, 21:00 | 2025-09-02 21:00:00 UTC | 2025-09-03 00:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 00:00 | 2025-09-03 00:00:00 UTC | 2025-09-03 03:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 03:00 | 2025-09-03 03:00:00 UTC | 2025-09-03 06:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 06:00 | 2025-09-03 06:00:00 UTC | 2025-09-03 09:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 09:00 | 2025-09-03 09:00:00 UTC | 2025-09-03 12:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 12:00 | 2025-09-03 12:00:00 UTC | 2025-09-03 15:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 15:00 | 2025-09-03 15:00:00 UTC | 2025-09-03 18:00:00 UTC | 0 | 0 | 0 |
| Sep 3, 18:00 | 2025-09-03 18:00:00 UTC | 2025-09-03 21:00:00 UTC | 1 | 0 | 1 |
What Hacker News says
3 sourcesImproving LLM token usage when debugging
6 points · Open discussionWhen debugging with an LLM, a failed build sends ~200 tokens of mostly useless output. The actual error? Maybe 60 tokens. Multiply that by 20-30 commands per debugging session, and you're burning through tokens like crazy.
So, I created a CLI tool that acts as a smart filter between your commands and the LLM. It knows what errors look like across different tech stacks and only shows what matters.
Before:
``` bash
> npm run build:graphql && react-router typegen && tsc && react-router build
> build:graphql
> graphql-codegen
Parse Configuration
Generate outputs
app/features/tasks/services/atoms.ts:55:60 - error TS2339: Property 'taskId' does not exist on type '{ request: UpdateTaskRequest; }'.
55 const response = await apiClient.updateTask(params.taskId, params.request);
Read the rest
~~~~~~
Found 1 error in app/features/tasks/services/atoms.ts:55
```
After:
```bash
$ aex frontend-build
app/features/tasks/services/atoms.ts(55,60): error TS2339: Property 'taskId' does not exist
Done
```
That's it. When the build succeeds? Just "Done" - literally 1 token instead of 200.
Have a look! The full article is here:
https://github.com/byme8/apparatus.exec/discussions/1