| name | three-dtiles-interactive-live |
| description | Run real Resonite Link `interactive` verification for this repo, centered on host-side startup, session-root binding attachment, stable stdout/stderr capture, and guided live rerun checks. Use when the user wants machine-level validation of interactive behavior instead of dry-run or unit tests, especially after host, DI, shutdown, input-binding, or rerun changes. |
3D Tiles Interactive Live
Use this skill for live interactive checks against a Windows-hosted Resonite Link session. Prefer unit tests first. Switch to this skill when the question depends on real session-root bindings, live rerun behavior, or host startup under the actual Windows dotnet.exe path.
Warning: this workflow creates and removes live 3DTilesLink Session ... roots and starts a long-running interactive process. Do not run cleanup while that process is still being used for the test sequence.
Workflow
-
Start interactive through scripts/run_interactive_live.py.
Use it when you need the process PID plus stable stdout/stderr log files.
-
Use scripts/cleanup_sessions.py before or after runs when you need an explicit packaged cleanup step.
-
Use the startup logs as the primary truth source.
After startup, drive the live fields from Resonite itself and confirm Search query changed, Search resolved, Run started, and Run completed from the captured stdout log.
-
Cover the standard interactive cases rather than a single hard-coded edit sequence.
The default coverage set is: initial run, relocation by a new search or distant coordinate target, range-only change, move within range, and move out of range. Use that set as the baseline unless the user asks for a narrower or broader scenario.
-
Choose the exact field edits on the spot.
Do not treat one fixed sequence as canonical. Send the specific Search / Latitude / Longitude / Range changes that best isolate the behavior you need to verify, while keeping each step attributable in the logs.
-
Read the reference doc when the run behaves unexpectedly.
Use references/workflow.md for the standard coverage matrix, log markers, failure modes, and the sequence rules that keep the test valid.
Scripts
-
scripts/run_interactive_live.py
Start one live interactive process on Windows, capture stdout/stderr, and wait until startup succeeds or fails.
-
scripts/cleanup_sessions.py
Run the packaged cleanup step for stale 3DTilesLink Session ... roots on the selected live session.
Output
Return the script JSON plus the key facts:
- listener port
- interactive PID
- stdout/stderr log paths
- whether startup reached
Interactive input bindings attached
- which standard coverage cases were exercised
- whether
Search resolved appeared for each requested query
- whether
Run completed appeared for each requested rerun
- whether the result came from the stable startup path plus manual sequential live edits
