القائمة
Metrics and analytics for measuring the health and effectiveness of the data flywheel
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/qashsolutions/denali --skill flywheel-metricsيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
Medicare coverage assistant that helps patients understand approval requirements and prevent denials
Workflow adaptations for healthcare provider practices using Denali for denial prevention and appeal management
Workflow and behavior adaptations for SHIP Medicare counselors using Denali as a case management tool
استنادا إلى تصنيف SOC المهني
عرض SKILL.md
| name | flywheel-metrics |
| description | Metrics and analytics for measuring the health and effectiveness of the data flywheel |
| version | 1.0.0 |
| triggers | ["analytics_review","flywheel_health_check"] |
Track the health and effectiveness of the data flywheel. Measure what matters, ignore vanity metrics.
Appeal success rate improvement over time. If the flywheel works, appeals generated with Denali should have a higher success rate as more outcome data flows in.
| Metric | Formula | Target (Month 3) | Target (Month 6) |
|---|---|---|---|
| Total outcomes collected | COUNT(appeal_outcomes) | 100 | 1,000 |
| Outcome response rate | responded / sent followups | 25% | 35% |
| Active SHIP counselors | Users WHERE role=counselor AND active < 30d | 10 | 30 |
| Provider practices | COUNT(provider_practices) | 1 | 3 |
| Success rate (overall) | approved / total outcomes | Baseline | +5% |
| Unique CPT+CARC combos | DISTINCT pairs in outcomes | 20 | 100 |
| Flywheel-informed appeals | Appeals where flywheel data was injected | 0% | 30% |
The system only injects flywheel data into Claude's prompt when:
Below these thresholds, the data is too sparse to be reliable.
| File | Purpose |
|---|---|
flywheel_metrics materialized view | Aggregated outcome data |
get_flywheel_context() RPC | Query flywheel data for specific CPT+CARC |
refresh_flywheel_metrics() RPC | Nightly refresh |
supabase/functions/refresh-flywheel-metrics/ | Cron edge function |
app/src/lib/learning.ts | getFlywheelContext(), buildFlywheelPromptInjection() |
app/src/lib/skills-loader.ts | Injects flywheel data into system prompt |