linshen
3126ae4530
feat(test): support image attachments in Basic/System tests ( #196 )
2026-07-17 23:25:52 +08:00
linshen
ecbe788909
fix(evaluation): format provider error details
2026-06-07 10:28:39 +08:00
linshen
466cf6ae48
refactor(core): harden image compare evaluation prompts
2026-04-04 10:07:43 +08:00
linshen
7cc7b01b1f
refactor(core): harden image result evaluation prompts
2026-04-04 09:18:27 +08:00
linshen
b60bfb41f9
feat(image): complete text2image evaluation closure
2026-04-03 21:47:52 +08:00
linshen
4b33dbbae2
feat(core): add multimodal text2image evaluation support
2026-03-28 15:18:24 +08:00
linshen
8fb35548ce
refactor(core): wrap builtin template evidence in json blocks
2026-03-26 21:07:11 +08:00
linshen
414ec28d7e
feat(evaluation): migrate compare rewrite protocol and add calibration artifacts
2026-03-22 14:28:53 +08:00
linshen
f49edee170
feat(core): add structured compare evaluation pipeline
2026-03-21 14:36:17 +08:00
linshen
aa8fe80953
test(e2e): refresh evaluation fixtures and stabilize image flows
2026-03-15 16:30:00 +08:00
linshen
8e11ee784f
test(evaluation): refresh unit, e2e, and VCR coverage
...
- expand core and ui regression coverage for analysis, result, and compare evaluation flows
- refresh text analysis and evaluation VCR fixtures after the workspace/result semantics change
- cover stale-state and analyze-created workspace behaviors in integration and e2e tests
2026-03-18 09:35:43 +08:00
linshen
2f0df2bb8c
refactor(core): redesign evaluation protocol and template builders
...
- replace original/optimized semantics with analysis, result, and compare request types
- centralize evaluation prompt construction in shared builders
- tighten rendered sample and evidence behavior tests for core evaluation service
2026-03-18 09:35:43 +08:00
linshen
df8c2c0bdc
fix(core): improve evaluation result parsing robustness
2026-02-10 17:03:23 +08:00
linshen
1a45e86b5e
feat(core): support user feedback in evaluation templates
2026-02-09 23:48:55 +08:00
linshen
8467b96263
feat(test): 实现智能测试系统并改进测试策略
...
- 实现智能测试系统，统一测试入口
- 完成 Pro 路由重构和 Image 模式评估测试
- 为 Image 模式添加 data-testid 支持
- 完善 E2E 测试并修复选择器策略
- 优化 Seedream、OpenAI 和 SiliconFlow 适配器
- 移除调试日志，保留错误和警告日志
- 重命名 Pro 模式子模式（system/user → multi/variable）
- 清理测试代码并改进稳定性
2026-01-10 11:23:00 +08:00
linshen
df52b32366
feat(evaluation): 重构评估响应结构，新增 patchPlan 精准修复功能
...
- 评估响应用 patchPlan 替代 issues，支持 insert/replace/delete 操作
- 新增 patch-plan 工具处理文本补丁应用
- 模板统一输出结构：score + improvements + patchPlan + summary
- UI 新增"立即替换"按钮，支持 diff 可视化展示
- usePromptOptimizer 新增 saveLocalEdit 保存本地编辑为新版本
- originalPrompt/testContent 改为可选，提升灵活性
2025-12-25 23:36:00 +08:00
linshen
8e1b00e2f7
feat(evaluation): 新增 prompt-only 和 prompt-iterate 评估类型
...
主要变更：
- 扩展 EvaluationType 联合类型，新增 prompt-only 和 prompt-iterate
- 新增 16 个评估模板（basic/pro × system/user × zh/en × prompt-only/prompt-iterate）
- 实现 Provider 路由模式，全局 EvaluationPanel 事件正确路由到各模式
- 新增 useEvaluationContext 和 useProContext 上下文注入
- Context Workspaces 通过 defineExpose 暴露 reEvaluateActive 和 openIterateDialog
- 模式/子模式切换时自动清理评估状态
- EvaluationPanel 标题支持新评估类型
- 新增评估服务单元测试
2025-12-23 21:43:00 +08:00
linshen