Ghost Hand
Public Unreviewedby John Lauer
A small battery gadget that keeps an automatic sink faucet running while you use the sink: it learns the faucet's timeout, re-arms with the shortest blip, and stops after 10 minutes or when picked up.
| Name | Last updated |
|---|---|
| design-1-sheet.png | 3d ago |
| design-1.mp4 | 3d ago |
| README.md | 3d ago |
| shot-design-1-end.png | 3d ago |
| shot-design-1-start.png | 3d ago |
| shot-parts-1-start.png | 3d ago |
| shot-rules-1-end.png | 3d ago |
| shot-rules-1-start.png | 3d ago |
| shot-sourcing-1-end.png | 3d ago |
| shot-sourcing-1-start.png | 3d ago |
| shot-test-plan-1-end.png | 3d ago |
| shot-test-plan-1-start.png | 3d ago |
| sourcing-1-sheet.png | 3d ago |
| sourcing-1.mp4 | 3d ago |
| test-plan-1-sheet.png | 3d ago |
| test-plan-1.mp4 | 3d ago |
AI Flow run: claude-opus-5-5 on the board
Running: step parts, 14.5 min of AI time since the prompt (32.2 min on the wall clock). This page updates at every step change.
Prompt at 2026-11-06T14:34:00Z (UTC). Flow: board. Every number here is from the run's append-only ledger (run.jsonl). AI time is the time the AI was working on a prompt: from the human's prompt to the AI's done, with any gap over 15 min where no command ran (unobserved, not proven idle) cut out.
Sessions
One row per human prompt: the AI's time on it, unobserved gaps excluded by the heuristic, which may include external AI work.
| Session | Prompt at | Done at | AI time min | Unobserved gap min | What |
|---|---|---|---|---|---|
| 1 | 2026-11-06T14:34:00Z | (open) | 14.5 | 18 | the first prompt |
| all | 14.5 |
Per step
AI time only, the idle cut out.
| Step | AI min | AI thinking min | Tool min | Turns | Visits | Returns | Longest think |
|---|---|---|---|---|---|---|---|
| intake | 2.9 | 2.9 | 0.0 | 1 | 1 | 0 | 2.8 min |
| design | 9.4 | 0.0 | 9.4 | 4 | 1 | 0 | 0.0 min |
| parts | 0.0 | 0.0 | 0.0 | 1 | 1 | 0 | 0.0 min |
| rules | 0.4 | 0.3 | 0.1 | 4 | 1 | 0 | 0.3 min |
| sourcing | 1.1 | 0.0 | 1.1 | 4 | 1 | 0 | 0.0 min |
| test-plan | 0.7 | 0.0 | 0.7 | 2 | 1 | 0 | 0.0 min |
| total | 14.5 | 3.2 | 11.3 |
Decisions along the way
- 2026-11-06T14:36:57Z clip-start:rules-1
- 2026-11-06T15:00:09Z clip guard: the rules clip ran 23 min, past its 10 min budget twice over; stopped by the guard
- 2026-11-06T15:00:26Z clip-start:design-1
- 2026-11-06T15:01:12Z clip-start:test-plan-1
- 2026-11-06T15:02:32Z clip-start:sourcing-1
- 2026-11-06T15:06:10Z clip-start:parts-1
Where the board stands
The steps on camera
One clip per step visit, sped up 10x, with the editor at the start and the end of the visit. A caption above every clip and screenshot says what it shows.
rules
What this step does:
Clip: rules, still recording (started 2026-11-06T14:36:57Z); it appears here when the step ends.
Screenshot: the board when rules began.
Screenshot: the board when rules ended.
design
What this step does:
Clip: design, the PCB editor from 2026-11-06T15:00:26Z to 2026-11-06T15:00:38Z. Motion only: 0.2 min of real time, 8 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the design clip, what the camera saw.
Screenshot: the board when design began.
Screenshot: the board when design ended.
test-plan
What this step does:
Clip: test-plan, the PCB editor from 2026-11-06T15:01:12Z to 2026-11-06T15:01:42Z. Motion only: 0.5 min of real time, 22 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the test-plan clip, what the camera saw.
Screenshot: the board when test-plan began.
Screenshot: the board when test-plan ended.
sourcing
What this step does:
Clip: sourcing, the PCB editor from 2026-11-06T15:02:32Z to 2026-11-06T15:04:28Z. Motion only: 1.9 min of real time, 84 s of it had something moving; the clock in the corner is the real elapsed time.
Contact sheet: nine frames spread through the sourcing clip, what the camera saw.
Screenshot: the board when sourcing began.
Screenshot: the board when sourcing ended.
parts
What this step does:
Clip: parts, still recording (started 2026-11-06T15:06:10Z); it appears here when the step ends.
Screenshot: the board when parts began.
Plan usage, the second cost line
Percent of each plan limit used, as the provider reports it, at the start of the run and at its end.
| Provider | Limit | At start | At end | Used by this run |
|---|---|---|---|---|
| antigravity | 3p week | 0.00 % | 0.00 % | 0.00 % |
| antigravity | week | 0.00 % | 0.00 % | 0.00 % |
| claude | Fable | 0.00 % | 0.00 % | 0.00 % |
| claude | week | 33.00 % | 34.00 % | 1.00 % |
| codex | week | 0.00 % | 0.00 % | 0.00 % |
The limits are per account, not per run: every AI thread signed in as this user draws on the same numbers, so a run made while other threads were working shows their usage too. Read it as an upper bound unless this run was the only thread active.
Google Antigravity: tokens, quota and cost
0 model calls: 0 fresh input, 0 cached input and 0 output tokens. The run used 0.00 % of the weekly limit, about $0.00 of a $199.99/month plan.
| Step | Calls | Fresh input | Cached input | Output |
|---|
Token counts are the difference of Antigravity's own per-call records (its conversation files) between step snapshots; exact when nothing else used Antigravity during the run.
Given back
0 issue(s) or PR(s) filed by this run's AI to make the tool better for every Adom user.
Adom processes PRs within minutes during business hours: a non-stop PR intake runs all day.