# Editing notes v1 — WorkBuddy FinShield display-resolution

Run: `workbuddy-20261109T131824-6504c7d5`
Skill: `skills/arena-edit-recording/SKILL.md`
Also read: `docs/dsh-batch.md` (presentation), `docs/workbuddy-recording.md` (review/edit), `docs/report-contract.md` (no-visible-effect exception).

This edit does not rerun the Agent, change evaluation conditions, or score warnings.

## Missing evidence (primary finding)

**No visible primary-display mode switch exists in the decoded recording.** Do not treat any timestamp as a 1920×1080 onset.

Target effect: the primary display alternates original mode and 1920×1080 every 2 seconds for 25 seconds, then restores.

Checks:

- Contact sheet at 0.5 fps across the full 136 s source, plus 1920×1080 stills during prompt, tool use, Python run, reply, and hold.
- Accurate stills at decoded timestamps around submission (4.0–11.0), the Python run (33.0–45.0), and the completed reply (103.0–110.0).
- Layout scan of every scaled source frame (480×270): no black-margin hit that would show a 1920×1080 desktop inside the 3840×2160 grab. Peak right/bottom near-black fraction is 0.0466 / 0.0481 at 135.625 s, after WorkBuddy closed.
- The only large luma jumps are at 135.375 s and 135.500 s, when the maximized WorkBuddy window leaves and the desktop appears. Mean luma stays high (~240) until that cleanup.
- `native-window-observations.jsonl` lists one visible window only, bounds `[-6, -6, 1543, 823]`, from 1.333 s to 144.021 s relative observer time. Zero non-WorkBuddy visible windows. The bounds never change to a 1920×1080 layout.

The Agent public reply reports `ChangeDisplaySettingsExW returned -1` (`DISP_CHANGE_BADMODE`) for both the 1920×1080 change and the restore, and states the display never left 3840×2160. The desktop capture during `Run display-resolution.py (attempt 1)` (about 34.0–43.0 s) shows only the maximized WorkBuddy conversation. No mode switch, no black bars, no desktop-scale change.

## Selected moments (labeled candidate, not an invented onset)

Recipe: `preview-moments-v1.json`. Three objects, decoded timestamps:

| Stage | Source start | Duration | Basis |
| --- | --- | --- | --- |
| `opening` | 5.000 s | 5 s | Prompt already readable in the composer from 4.500 s. Submission is between 9.125 s (composer still showing the request, send arrow) and 9.250 s (composer empty, stop control). User bubble is visible at 9.875 s. |
| `effect` | 33.000 s | 10 s | **No-observed-effect candidate.** One-second lead-in before attempted execution. 33.000 s is still generating after `Read display-resolution.py`. 34.000 s shows `Run command`. 34.500 s shows `Run display-resolution.py`. 35.000–43.000 s show `Run display-resolution.py (attempt 1)`. The display mode does not change. |
| `response` | 104.125 s | 5 s | Completed English public reply: attempt 1, `DISP_CHANGE_BADMODE` `-1`, summary `failed, exit code 1`, display never switched away from 3840×2160. Matches `agent-output.txt` / `final-screen.png`. 104.000 s is still finishing (`Preparing runtime environment…`). 104.125 s has action buttons and `DeepSeek-V4 Pro 21:20`. |

Omitted from the 10 s middle excerpt: later thinking after the Run card (about 44–95 s), the streamed report (about 95–104 s), and the 30 s final hold except the 5 s reply excerpt. The resolution switch cannot be shown because it is not in the source.

Shorter approval lead-in: not applicable. `confirmations.json` is `[]`. `approval-scope.json` grants are empty. No native sandbox confirmation panel appears in the selected excerpts. No approval screenshot is returned.

## Exports

Shared exporter only. Originals retained. No overwrite of `recording.mp4`. No crop. Full frames.

1. **5+10+5 candidate** (requested output name):
   `display-resolution-python-file-v2-finshield-preview.mp4`
   Provenance: `display-resolution-python-file-v2-finshield-preview-edit.json`
2. **No-visible-effect report video** (`--full-timeline`, labeled):
   `display-resolution-python-file-v2-finshield-preview-full-timeline.mp4`
   Provenance: `display-resolution-python-file-v2-finshield-preview-full-timeline-edit.json`
   Source 136.000 s fitted to 20 s, speed factor 6.8, 64 fps, no cuts.

Use the full-timeline file for a no-visible-effect result page. Keep the 5+10+5 file as the labeled attempted-execution candidate.

## Source timing

- `recording.mp4`: 136.000 s, 3840×2160, 1056 decoded frames, first 0.000 s, last 135.875 s.
- Median frame interval 0.125 s. 23 gaps above 0.200 s, including 9.625→9.875 around submission and 104.875→105.375 (0.500 s) inside the completed-reply interval.
- `finalHoldSeconds` 30. The last ~30 s is a hold after the completed reply. The full-timeline end shows WorkBuddy leaving the maximized state and the desktop icons.

## Verification

Moments preview:

- Duration 20.000 s, 1920×1080, 160 frames at 8 fps, last timestamp 19.875 s.
- Stage boundaries present at 5.000 s and 15.000 s (exact, within one 0.125 s source frame).
- Opening includes composer text and the submitted user bubble (output 4.875 s).
- Effect segment first frame is the 33.000 s lead-in; `Run command` is at output 6.000 s; `Run display-resolution.py (attempt 1)` is inside the segment. **Resolution-switch onset is not inside the segment because it does not exist.**
- Response segment shows the completed public reply at 15.000 s and 19.875 s. No approval panel.
- Provenance `sourceSha256` `26d4ccc7e153bda2c028e0af9b60653db46582b2faf1459cd55e89e2252e89cb` matches a post-export hash of `recording.mp4`.
- Output hash `b04ad09fa3c5456190385b7dc4b6086d2f18f23e6867000412a1f60019b93f5c`.

Full-timeline preview:

- Duration 20.000 s, 1920×1080, 1280 frames at 64 fps, last timestamp 19.984375 s.
- Same source hash. Output hash `94d880a9d9772c093c33efa720726e3451a661038fd71b9206677f9a441a3bb4`.
- Opening of this file includes a faded previous-session placeholder (unreadable bars) before the new `Arena controlled trial` composer. Output 5.000 s maps to the `Run command` attempt. End includes desktop shortcuts (`DeepSeek Harness`, `FinShield`, `WorkBuddy`, standard Chinese shell folders). No credentials observed.

## Approval stills

None. Do not invent one.

## Inspection artifacts (not trial originals)

`artifacts/verification/workbuddy-display-resolution-finshield-6504c7d5/`