科研 Skill
PPT 阅读器
ppt-reader
读取本地 PPT 文本、表格和备注
Skill 说明
读取和分析本地 PowerPoint PPTX/POTX 演示文稿,提取幻灯片文本、表格和演讲者备注;在 office 渲染工具可用时可生成页面图像用于视觉检查 | 输入:.pptx / .potx 文件 | 输出:Markdown 文本、表格、备注和可选页面渲染图
SKILL.md
PPT Reader
Use this skill for presentation-reading requests. A PPTX/POTX is an OOXML ZIP package.
Default text workflow
Use MarkItDown by default. It is a Python package, so install it automatically when it is missing; do not ask the user to install it manually.
Resolve PYTHON in this order: $CODAR_PYTHON, then python3, then python. Keep the package in this skill's private .venv, never in the user's system Python:
PYTHON="${CODAR_PYTHON:-python3}"
if [ ! -x "$SKILL_DIR/.venv/bin/markitdown" ]; then
"$PYTHON" -m venv "$SKILL_DIR/.venv"
"$SKILL_DIR/.venv/bin/python" -m pip install --disable-pip-version-check "markitdown[pptx]"
fi
"$SKILL_DIR/.venv/bin/markitdown" presentation.pptx
Use the extracted Markdown to answer questions, and retain slide-number markers when they are present. If the MarkItDown download or conversion fails, continue with the bundled standard-library fallback instead of failing the request:
"${CODAR_PYTHON:-python3}" "$SKILL_DIR/scripts/read_pptx.py" presentation.pptx
The fallback extracts ordered slide text, table cells, and speaker notes without non-standard Python packages. It does not understand pictures, charts, SmartArt, or visual layout.
Visual workflow
Check first:
command -v soffice
command -v pdftoppm
Only if both commands exist, render for visual inspection:
soffice --headless --convert-to pdf --outdir ./pptx-render presentation.pptx
pdftoppm -png -r 144 ./pptx-render/presentation.pdf ./pptx-render/slide
If either command is absent, clearly state that text was extracted but visual inspection was skipped. Do not require LibreOffice or Poppler for ordinary text questions.
Limits
.pptis the legacy binary format; ask for a.pptxexport.- Do not claim that a chart, screenshot, or diagram says something based only on text extraction. Render slides first when visual evidence matters.
相关科研 Skill
中药材分子鉴定用ITS2等DNA条形码鉴定药材基原物种,处理多基原与降解样品局限。适用于药材真伪鉴别研究,不替代法定检验。
中药网络药理学用TCMSP等数据库构建成分-靶点-通路网络并做富集分析,批判性讨论方法学局限。适用于中药机制假设生成与论文方法复核。
中医术语标准与知识图谱建立中医药术语与WHO/ICD-11标准的映射,设计本体并用Neo4j构建知识图谱。适用于数据整合与科研检索,不作临床决策系统。
方剂配伍规律挖掘对方剂做药名规范化、关联规则与网络聚类分析,发现候选核心药组。适用于用药规律挖掘,结果仅为待验证假设,不作处方依据。
证候与症状数据分析对症状量表数据做术语规范、潜在类别与因子分析,探索统计学证候模式。适用于证候分型研究,结果不用于个体辨证诊断。
泵用混合物热力学计算Calculates thermodynamic properties and mixture phase behavior for pump applications using the thermo Python library’s DIPPR/NIST correlations. Use for VLE or mixture property calculations.
常见问题
PPT 阅读器能做什么?
读取和分析本地 PowerPoint PPTX/POTX 演示文稿,提取幻灯片文本、表格和演讲者备注;在 office 渲染工具可用时可生成页面图像用于视觉检查 | 输入:.pptx / .potx 文件 | 输出:Markdown 文本、表格、备注和可选页面渲染图。
如何安装PPT 阅读器?
在科搭 Codar 左侧点击「插件」,在「组件」页签搜索「PPT 阅读器」并安装;在 Claude Code 或 Codex 中可以发送本页的安装提示词,或下载 Skill 包后解压到 ~/.claude/skills 或 ~/.codex/skills。
PPT 阅读器需要哪些权限?
该科研 Skill声明需要:本地文件。安装前会再次列出,由你确认后才会生效。
PPT 阅读器经过官方审核吗?
是。SpeedSkills 收录的每个版本都会记录内容摘要与权限声明,当前版本已通过安全审核。