一键导入
verify
Background knowledge for droid-control workflows -- not invoked directly. Deliverable verification against commitments.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Background knowledge for droid-control workflows -- not invoked directly. Deliverable verification against commitments.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | verify |
| description | Background knowledge for droid-control workflows -- not invoked directly. Deliverable verification against commitments. |
| user-invocable | false |
The orchestrator routed you here. This atom checks the final deliverable against the commitments made at the start of the workflow.
You receive:
Run ffprobe on the final .mp4:
ffprobe -v quiet -print_format json -show_format -show_streams <video>
| Check | Pass condition |
|---|---|
| Exists and plays | ffprobe exits 0, duration > 0 |
| Resolution | 1920x1080 (or target resolution) |
| Pixel format | yuv420p |
| File size | Under 5 MB for GitHub embeds (25 MB hard limit) |
| Duration | Must fall within the compose pacing table's target range for this demo type: 30-45s (single feature), 45-75s (side-by-side comparison), 60-120s (multi-phase). Below the minimum is a failure — re-compose with a lower speed factor or re-capture with more steps. |
| Filename | Includes PR number or meaningful identifier |
Walk through each commitment from the parse step:
| Commitment | How to verify |
|---|---|
| Title card | Video starts with a static frame showing PR info (check first 5s) |
| Side-by-side layout | Video shows two panels with a divider |
| Showcase polish | Resolution is 1920x1080, window chrome and rounded corners visible |
| Keystroke overlay | Pill overlays appear at interaction points |
| Effects | Effects matching the committed tier are present (utilitarian: zoom/keystroke; full: spotlight, zoom, callout, keystroke) |
| Speed note | Title card mentions playback speed |
| Check | Pass condition |
|---|---|
| Evidence exists | Screenshots/snapshots at every claimed proof point |
| Environment stated | Driver, terminal/browser, OS identified |
| Conclusion present | Evidence explicitly supports or refutes the claim |
| Before/after paired | If comparison, both branches shown at same capture points |
| Check | Pass condition |
|---|---|
| Step coverage | Every defined test step has a result (PASS/FAIL) |
| Evidence attached | Screenshots/snapshots at every step |
| Failures documented | Failed steps have evidence and description |
| Report structured | Markdown report follows the QA template |
If any check fails:
Do not report a deliverable as complete until every commitment is met.
## Verification
### Technical
- Resolution: 1920x1080 ✓
- Duration: 42s ✓
- Size: 3.2 MB ✓
- Format: yuv420p ✓
### Commitments
- [x] Title card with PR info
- [x] Side-by-side comparison layout
- [x] Showcase hero preset applied
- [x] Keystroke overlay visible
### Content
- [x] Fork creates independent session (visible at 0:18)
- [x] History diverges after fork (visible at 0:32)
All commitments met. Deliverable ready.
Background knowledge for droid-control workflows -- not invoked directly. Recording lifecycle for terminal and browser sessions.
Background knowledge for droid-control workflows -- not invoked directly. Desktop-control driver mechanics for native GUI app automation via trycua cua-driver.
Control terminal TUIs and web/Electron apps for testing, demos, QA, and computer-use tasks. Use when you need to automate a CLI, drive a browser, record a demo, or capture proof artifacts.
Create a pull request with Conventional Commits formatting, a templated body, and local verification. Use when the user asks to create a PR, open a PR, submit changes for review, or put code up for review.
Follow up on an existing PR by rebasing on the base branch, addressing reviewer comments, fixing CI issues, and pushing updates. Use when the user provides a PR URL or number and wants to get it ready for merge.
Intercept and debug HTTP traffic from any CLI, service, or script using HTTP Toolkit. Use when you need to inspect LLM API calls, backend requests, auth flows, or debug network-level issues across any language or runtime.