From 9e2c02a3337cc8789c18cb763fc1ddc28a0d03a4 Mon Sep 17 00:00:00 2001
From: steven_li <liyuxuan2001@hotmail.com>
Date: Mon, 8 Jun 2026 13:38:10 +0800
Subject: [PATCH] feat(skills-ui): show replay eval coverage

---
 .../backend/beaver/interfaces/web/app.py        |  9 +++++++--
 app-instance/frontend/app/(app)/skills/page.tsx | 17 +++++++++++++++++
 app-instance/frontend/types/index.ts            |  9 +++++++++
 3 files changed, 33 insertions(+), 2 deletions(-)

diff --git a/app-instance/backend/beaver/interfaces/web/app.py b/app-instance/backend/beaver/interfaces/web/app.py
index 3ef6f11..9b88f86 100644
--- a/app-instance/backend/beaver/interfaces/web/app.py
+++ b/app-instance/backend/beaver/interfaces/web/app.py
@@ -50,6 +50,7 @@ from beaver.services.user_file_resolver import (
     build_file_auth_context,
 )
 from beaver.skills.learning import SkillLearningWorker, SkillLearningWorkerConfig
+from beaver.skills.learning.replay import ReplayRunner
 from beaver.skills.catalog.utils import parse_frontmatter
 
 from .deps import get_agent_service
@@ -2080,7 +2081,8 @@ def create_app(
     @app.post("/api/skills/candidates/{candidate_id}/draft")
     async def synthesize_skill_draft(candidate_id: str, request: Request) -> dict[str, Any]:
         agent_service = get_agent_service(request)
-        loaded = agent_service.create_loop().boot()
+        loop = agent_service.create_loop()
+        loaded = loop.boot()
         try:
             candidate = loaded.skill_learning_pipeline.get_candidate(candidate_id)  # type: ignore[union-attr]
             if candidate.draft_skill_name and candidate.draft_id:
@@ -2099,6 +2101,7 @@ def create_app(
                 draft.skill_name,
                 draft.draft_id,
                 provider_bundle=provider_bundle,
+                replay_runner=ReplayRunner(agent_loop=loop),
             )
         except ValueError as exc:
             raise HTTPException(status_code=404, detail=str(exc)) from exc
@@ -2107,7 +2110,8 @@ def create_app(
     @app.post("/api/skills/candidates/{candidate_id}/regenerate")
     async def regenerate_skill_draft(candidate_id: str, request: Request) -> dict[str, Any]:
         agent_service = get_agent_service(request)
-        loaded = agent_service.create_loop().boot()
+        loop = agent_service.create_loop()
+        loaded = loop.boot()
         provider_bundle = agent_service._make_provider_bundle_for_task(loaded, {})  # noqa: SLF001
         try:
             draft = await loaded.skill_learning_pipeline.regenerate_draft(  # type: ignore[union-attr]
@@ -2120,6 +2124,7 @@ def create_app(
                 draft.skill_name,
                 draft.draft_id,
                 provider_bundle=provider_bundle,
+                replay_runner=ReplayRunner(agent_loop=loop),
             )
         except ValueError as exc:
             raise HTTPException(status_code=404, detail=str(exc)) from exc
diff --git a/app-instance/frontend/app/(app)/skills/page.tsx b/app-instance/frontend/app/(app)/skills/page.tsx
index 6d23e35..e46896f 100644
--- a/app-instance/frontend/app/(app)/skills/page.tsx
+++ b/app-instance/frontend/app/(app)/skills/page.tsx
@@ -1088,6 +1088,12 @@ function EvalReportPanel({ report }: { report?: SkillDraftEvalReport | null }) {
         />
       </div>
 
+      <div className="mt-3 grid gap-2 sm:grid-cols-3">
+        <MetricTile label={t('执行覆盖', 'Execution')} value={formatPercent(report.execution_coverage)} />
+        <MetricTile label={t('替代评估', 'Surrogate')} value={formatPercent(report.surrogate_coverage)} />
+        <MetricTile label={t('置信度', 'Confidence')} value={report.confidence || 'low'} />
+      </div>
+
       <div className="mt-3 grid gap-2 sm:grid-cols-3">
         <ReadableFact icon={<CheckCircle2 className="h-4 w-4" />} label={t('改进', 'Improved')} value={String(report.improved_count)} />
         <ReadableFact icon={<XCircle className="h-4 w-4" />} label={t('回退', 'Regressed')} value={String(report.regression_count)} />
@@ -1135,6 +1141,12 @@ function EvalReportPanel({ report }: { report?: SkillDraftEvalReport | null }) {
           </div>
         </div>
       )}
+      {Array.isArray(report.case_reports) && report.case_reports.length > 0 ? (
+        <RawDetails title={t('Replay case reports', 'Replay case reports')} payload={report.case_reports} />
+      ) : null}
+      {report.preservation_report ? (
+        <RawDetails title={t('Preservation report', 'Preservation report')} payload={report.preservation_report} />
+      ) : null}
       <div className="mt-3 text-xs text-muted-foreground">{formatDateTime(report.created_at)}</div>
       <RawDetails title={t('原始评估报告', 'Raw eval report')} payload={report} />
     </div>
@@ -1387,6 +1399,11 @@ function formatScore(value: number): string {
   return value.toFixed(2);
 }
 
+function formatPercent(value?: number | null): string {
+  if (typeof value !== 'number' || Number.isNaN(value)) return '0%';
+  return `${Math.round(value * 100)}%`;
+}
+
 function formatSignedScore(value: number): string {
   if (!Number.isFinite(value)) return '-';
   return `${value >= 0 ? '+' : ''}${value.toFixed(2)}`;
diff --git a/app-instance/frontend/types/index.ts b/app-instance/frontend/types/index.ts
index c0ae12c..5db526c 100644
--- a/app-instance/frontend/types/index.ts
+++ b/app-instance/frontend/types/index.ts
@@ -985,6 +985,15 @@ export interface SkillDraftEvalReport {
   cases: Array<Record<string, unknown>>;
   status: string;
   created_at: string;
+  eval_version?: string;
+  mode?: 'heuristic' | 'replay' | string;
+  execution_coverage?: number;
+  surrogate_coverage?: number;
+  blocked_coverage?: number;
+  confidence?: 'low' | 'medium' | 'high' | string;
+  case_reports?: Array<Record<string, unknown>>;
+  tool_mode_summary?: Record<string, unknown>;
+  preservation_report?: Record<string, unknown> | null;
 }
 
 export interface SkillDraft {