{"id":40993,"plugin_id":"plugin_asdk_app_6a53f7d79b5c81919a02e9814eb9f070","kind":"skill","collection_source":"plugin_package","comparison_source":null,"observed_at":"2026-10-24T12:56:56.271Z","digest":"4a01d2c6ba4fc4f33814f871d03ef39a66a0affbcc042fe44a53cea59a3242ac","against":null,"payload":{"description":"Reviews Supademo OS, browser, and country view shares and identifies demos worth checking on mobile. Use for a mobile campaign or accessibility review; device-specific completion rates are unavailable from these tools.","included_files":[{"relative_path":"agents/openai.yaml","size_in_bytes":217}],"name":"device-experience-auditor","skill_md_contents":"---\nname: device-experience-auditor\ndescription: Reviews Supademo OS, browser, and country view shares and identifies demos worth checking on mobile. Use for a mobile campaign or accessibility review; device-specific completion rates are unavailable from these tools.\n---\n\n# Device Experience Auditor\n\nUse audience data to prioritize a mobile playback review without inventing device conversion metrics.\n\n## Workflow\n\n1. **Set the scope.** Select the workspace and UTC period. Pass `viewType` explicitly. `get_device_analytics` without `demoId` combines demo and showcase traffic; pass a demo ID for a demo-specific view.\n2. **Read the returned breakdowns.** Report OS and browser view shares and their `truncation` flags. These are top-five lists, so missing entries can be in the long tail. Country shares exclude unknown locations and need a coverage caveat.\n3. **State the measurement limit.** The tool does not return device classes, per-device completion, or a mobile-versus-desktop conversion gap. OS names can suggest mobile traffic, but label that interpretation and avoid an exact mobile total when the list is truncated or ambiguous.\n4. **Inspect candidates.** Use `get_demo` to review screenshot density, annotation length, and hotspot layout. If browser access is available, check actual playback at a mobile viewport and a desktop viewport. Distinguish observed layout problems from hypotheses based on metadata.\n5. **Recommend priorities.** Rank by measured traffic and observed usability problems. Suggest a shorter duplicate, clearer annotations, or recapture at a useful zoom. Do not calculate a traffic-times-completion-gap score when completion by device is unavailable.\n6. **Report.** For each demo, include the period, audience scope, observed OS/browser shares, truncation, playback evidence, and recommended action. Keep audit-only requests read-only; carry out fixes when the user has requested them.\n\n## Supademo tool conventions\n\nRead [the tool conventions](../supademo-onboarding/references/tool-use.md) before calling Supademo tools for this workflow.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}