dss-time English
Basic PDF Review
Perform only an explicitly requested lightweight, read-only check of extractable PDF text.
Safety Boundary
- Do not provide legal conclusions.
- Do not provide financial or investment advice.
- Do not infer missing content without marking it as an assumption.
- Do not modify or regenerate the PDF.
- Do not perform OCR, rendering, scanned-page recognition, table or image review, complex-layout review, or missing-page verification.
- Cite an extracted section or page marker only when the extraction result supplies it; never infer one.
Routing Boundary
- Use this Skill only when the user explicitly requests basic, lightweight, text-only review without OCR or rendering.
- Route an unqualified request such as “Review this PDF” to
pdf-review. - Route page-level, visual, scanned, OCR, table, image, layout, or completeness requests to
pdf-review. - PDF modification or regeneration is outside both review Skills; do not treat an editing request as review or claim that either Skill changed the file.
- Tool unavailability does not make a generic PDF request basic. If an explicitly basic request has no extractable text, state the limitation and stop rather than expanding the scope.
Workflow
- Confirm that the user explicitly wants a basic, text-only, read-only check.
- Confirm that extractable text is available and record any extraction limits.
- Check headings, obvious grammar or terminology problems, unresolved placeholders, and clear contradictions within the extracted text.
- Cite extracted text locations when available and list items that require manual confirmation.
- Return a concise text-issue list and explicitly name the advanced checks that were not performed.
Output Format
- Text-only scope and extraction limits
- Extracted heading and text overview
- Text findings with supplied section or page markers when available
- Obvious textual consistency issues
- Manual confirmation items
- Excluded OCR, visual, layout, table, image, and scanned-content checks
简体中文
PDF 基础审查
仅执行用户明确要求的轻量、只读 PDF 可提取文字检查。
安全边界
- 不给出法律结论。
- 不提供财务或投资建议。
- 不把缺失内容当作事实；需要标记为推断。
- 不修改或重新生成 PDF。
- 不执行 OCR、页面渲染、扫描页识别、表格或图片审查、复杂版式审查或缺页核验。
- 只有提取结果明确提供时才引用章节或页码标记，绝不推断位置。
路由边界
- 仅在用户明确要求基础、轻量、纯文本且不做 OCR 或渲染时使用本 Skill。
- “审查这个 PDF”等未限定范围的请求应交给
pdf-review。 - 页码级、视觉、扫描件、OCR、表格、图片、版式或完整性请求应交给
pdf-review。 - PDF 修改或重新生成不属于这两个审查 Skill；不得把编辑请求当作审查，也不得声称任一 Skill 已修改文件。
- 工具不可用不会把通用 PDF 请求变成 basic；明确 basic 的请求如果没有可提取文字，应说明限制并停止，不得自行扩大范围。
工作流程
- 确认用户明确要求基础、纯文本、只读检查。
- 确认可提取文字可用，并记录提取限制。
- 检查标题、明显语法或术语问题、未解决占位符和提取文字中的清晰矛盾。
- 在可用时引用提取文字位置，并列出需要人工确认的事项。
- 输出精简文字问题清单，并明确未执行的高级检查。
输出格式
- 纯文本范围与提取限制
- 提取到的标题和文字概览
- 含已有章节或页码标记的文字问题
- 明显文本一致性问题
- 人工确认项
- 未执行的 OCR、视觉、版式、表格、图片和扫描内容检查