How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

wow-viewer World Runtime 01 Negative Asset Lookup Suppression - Prompt

Name: wow-viewer World Runtime 01 Negative Asset Lookup Suppression
Author: akspa0

Prompt file imported from akspa0/parp-tools (.codex/prompts/wow-viewer-world-runtime/01-negative-asset-lookup-suppression.md). Copyright stays with the author.

Implement or plan the first stabilization slice for the WorldScene split: stop repeated missing asset work from distorting runtime behavior and logs.

Read First

gillijimproject_refactor/memory-bank/activeContext.md
gillijimproject_refactor/memory-bank/progress.md
gillijimproject_refactor/plans/wow_viewer_world_runtime_service_plan_2026-03-31.md
gillijimproject_refactor/src/MdxViewer/Terrain/WorldAssetManager.cs
gillijimproject_refactor/src/MdxViewer/Rendering/WmoRenderer.cs
gillijimproject_refactor/src/MdxViewer/ViewerApp.cs
gillijimproject_refactor/src/MdxViewer/ViewerApp_CaptureAutomation.cs
wow-viewer/README.md
AGENTS.md

Goal

Remove obvious repeated asset-miss work, especially .skin companion lookups and failed MDX retries, before larger runtime-service extraction starts.

Current Concrete Problem

WorldAssetManager currently retries cached failed MDX loads and deferred failed MDX loads.
M2-family loads still walk candidate .skin paths even when the practical answer is already known to be "no usable companion skin exists for this model path".
similar companion-skin loops exist in standalone load paths and WMO doodad M2 loading.
this creates log spam, repeated file probes, and likely frame-time noise that makes broader renderer work harder to measure accurately.

Non-Negotiable Constraints

Do not turn this slice into the full WorldScene service split.
Do not over-abstract asset loading just to match a future runtime design.
If a reusable negative-cache contract belongs in WowViewer.Core.Runtime, move only that narrow seam.
If the fastest correct fix needs to stay in MdxViewer first, do that and describe it honestly as compatibility prep.
Do not claim renderer-performance closure from reducing one miss path.

What The Work Must Produce

the exact repeated asset-miss paths to stop
the exact files that should own negative-cache or failure-residency behavior
the exact logging changes needed so the app reports the problem once without flooding
a narrow proof plan using build, tests where practical, and optional fixed-shot capture smoke
explicit follow-on hooks for slice 02 instead of another broad cleanup bucket

Install wow-viewer World Runtime 01 Negative Asset Lookup Suppression and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

Implement or plan the first stabilization slice for the `WorldScene` split: stop repeated missing asset work from distorting runtime behavior and logs. ## Read First 1. `gillijimproject_refactor/memory-bank/activeContext.md` 2. `gillijimproject_refactor/memory-bank/progress.md` 3. `gillijimproject_refactor/plans/wow_viewer_world_runtime_service_plan_2026-03-31.md` 4. `gillijimproject_refactor/src/MdxViewer/Terrain/WorldAssetManager.cs` 5. `gillijimproject_refactor/src/MdxViewer/Rendering/WmoRenderer.cs` 6. `gillijimproject_refactor/src/MdxViewer/ViewerApp.cs` 7. `gillijimproject_refactor/src/MdxViewer/ViewerApp_CaptureAutomation.cs` 8. `wow-viewer/README.md` 9. `AGENTS.md` ## Goal Remove obvious repeated asset-miss work, especially `.skin` companion lookups and failed MDX retries, before larger runtime-service extraction starts. ## Current Concrete Problem - `WorldAssetManager` currently retries cached failed MDX loads and deferred failed MDX loads. - M2-family loads still walk candidate `.skin` paths even when the practical answer is already known to be "no usable companion skin exists for this model path". - similar companion-skin loops exist in standalone load paths and WMO doodad M2 loading. - this creates log spam, repeated file probes, and likely frame-time noise that makes broader renderer work harder to measure accurately. ## Non-Negotiable Constraints - Do not turn this slice into the full `WorldScene` service split. - Do not over-abstract asset loading just to match a future runtime design. - If a reusable negative-cache contract belongs in `WowViewer.Core.Runtime`, move only that narrow seam. - If the fastest correct fix needs to stay in `MdxViewer` first, do that and describe it honestly as compatibility prep. - Do not claim renderer-performance closure from reducing one miss path. ## What The Work Must Produce 1. the exact repeated asset-miss paths to stop 2. the exact files that should own negative-cache or failure-residency behavior 3. the exact logging changes needed so the app reports the problem once without flooding 4. a narrow proof plan using build, tests where practical, and optional fixed-shot capture smoke 5. explicit follow-on hooks for slice 02 instead of another broad cleanup bucket

{ "prompt": "Implement or plan the first stabilization slice for the `WorldScene` split: stop repeated missing asset work from distorting runtime behavior and logs.\n\n## Read First\n\n1. `gillijimproject_refactor/memory-bank/activeContext.md`\n2. `gillijimproject_refactor/memory-bank/progress.md`\n3. `gillijimproject_refactor/plans/wow_viewer_world_runtime_service_plan_2026-03-31.md`\n4. `gillijimproject_refactor/src/MdxViewer/Terrain/WorldAssetManager.cs`\n5. `gillijimproject_refactor/src/MdxViewer/Rendering/WmoRenderer.cs`\n6. `gillijimproject_refactor/src/MdxViewer/ViewerApp.cs`\n7. `gillijimproject_refactor/src/MdxViewer/ViewerApp_CaptureAutomation.cs`\n8. `wow-viewer/README.md`\n9. `AGENTS.md`\n\n## Goal\n\nRemove obvious repeated asset-miss work, especially `.skin` companion lookups and failed MDX retries, before larger runtime-service extraction starts.\n\n## Current Concrete Problem\n\n- `WorldAssetManager` currently retries cached failed MDX loads and deferred failed MDX loads.\n- M2-family loads still walk candidate `.skin` paths even when the practical answer is already known to be \"no usable companion skin exists for this model path\".\n- similar companion-skin loops exist in standalone load paths and WMO doodad M2 loading.\n- this creates log spam, repeated file probes, and likely frame-time noise that makes broader renderer work harder to measure accurately.\n\n## Non-Negotiable Constraints\n\n- Do not turn this slice into the full `WorldScene` service split.\n- Do not over-abstract asset loading just to match a future runtime design.\n- If a reusable negative-cache contract belongs in `WowViewer.Core.Runtime`, move only that narrow seam.\n- If the fastest correct fix needs to stay in `MdxViewer` first, do that and describe it honestly as compatibility prep.\n- Do not claim renderer-performance closure from reducing one miss path.\n\n## What The Work Must Produce\n\n1. the exact repeated asset-miss paths to stop\n2. the exact files that should own negative-cache or failure-residency behavior\n3. the exact logging changes needed so the app reports the problem once without flooding\n4. a narrow proof plan using build, tests where practical, and optional fixed-shot capture smoke\n5. explicit follow-on hooks for slice 02 instead of another broad cleanup bucket", "variables": [], "notes": "Implement or plan the first world-runtime stabilization slice for wow-viewer and MdxViewer. Use when repeated .skin lookups, failed MDX retry spam, negative asset misses, or noisy asset logs are causing hidden performance loss before the deeper WorldScene split. Arguments: Optional asset family, hotspot file, or missing-lookup symptom to prioritize.", "metadata": { "source": { "provider": "github-codex-prompts", "repository": "https://github.com/akspa0/parp-tools", "path": ".codex/prompts/wow-viewer-world-runtime/01-negative-asset-lookup-suppression.md", "ref": "5608d59a180dc4296e29bc19be1a3cc5e94ef309", "url": "https://github.com/akspa0/parp-tools/blob/5608d59a180dc4296e29bc19be1a3cc5e94ef309/.codex/prompts/wow-viewer-world-runtime/01-negative-asset-lookup-suppression.md", "key": "akspa0/parp-tools/.codex/prompts/wow-viewer-world-runtime/01-negative-asset-lookup-suppression.md" } } }

wow-viewer World Runtime 01 Negative Asset Lookup Suppression

About

Read First

Goal

Current Concrete Problem

Non-Negotiable Constraints

What The Work Must Produce

Install wow-viewer World Runtime 01 Negative Asset Lookup Suppression and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews