Analyze facial/body photos to detect signs of cosmetic surgery or aesthetic procedures. Use when the user uploads a photo and asks to identify cosmetic work, detect plastic surgery, assess facial naturalness, check if someone has had work done, analyze before/after photos, or evaluate aesthetic procedure signs. Also trigger when users ask about specific procedures visible in photos (fillers, Botox, rhinoplasty, jaw contouring, etc.), compare photos for surgical changes, or want a "naturalness score" for a face. Works with single images, multiple comparison images, and video screenshots. 触发词：整容检测、看看有没有整、自然度评分、鉴定一下、有没有动过。
Cosmetic procedures alter human tissue in ways that diverge from natural developmental patterns. Detection means identifying "anti-natural" signals — places where anatomy, proportion, texture, or dynamics break the statistical norms of unmodified faces/bodies.
This is adversarial: the best work is designed to be undetectable. Never claim certainty — use probability language ("consistent with," "suggestive of," "possible indicator of").
Before analysis, read references/analysis-framework.md for the detailed region-by-region indicator checklist.
Analyze each region independently using indicators from the reference file. For each region assess:
The most powerful detection layer. Natural faces have internal consistency. Look for:
## 整容检测分析 / Cosmetic Procedure Detection Analysis
### 基础信息 / Baseline
- 图像质量评估 / Image quality assessment
- 参考人种基线 / Ethnic baseline reference
- 年龄估计 / Estimated age
- 滤镜/修图评估 / Filter/editing assessment
### 区域分析 / Regional Analysis
For each region with findings:
- 观察到的特征 / Observed features
- 可能的项目 / Possible procedure(s)
- 置信度 / Confidence: Low|Medium|High
- 判断依据 / Reasoning
### 整体协调性 / Cross-Region Coherence
- 种族特征一致性 / Ethnic feature consistency
- 年龄一致性 / Age consistency
- 对称性分析 / Symmetry analysis
### 总评 / Overall Assessment
- 自然度评分 / Naturalness score (1-10, 10=completely natural)
- 最可能的项目清单 / Most likely procedures (if any)
- 整体置信度 / Overall confidence
- 重要声明 / Important disclaimer
Use the user's language. Template above is bilingual for reference.
When 2+ photos of the same person at different times are provided:
When analyzing multiple people (group photo, set of photos):