メニュー
Convert a PDF file to Markdown. Use this when the user wants to convert a PDF to markdown, extract text from a PDF, or get markdown content from a PDF file.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Adversarial counterargument posture. Use when the user wants the strongest possible case AGAINST a recent decision, claim, architecture, or plan — a deliberate dose of disagreement to surface blind spots and confirmation bias. Pairs well with /think-with-me.
Prior-art investigator posture. Use before architecture decisions to gather what's already been tried, written about, or built. Produces an annotated bibliography rather than a synthesis. Pairs well with /think-with-me.
Socratic challenger posture. Use when the user brings a fuzzy idea ("build me X", "should we use Y?", "I'm thinking about Z") and would benefit more from being challenged than from immediate execution. Activates a no-code-until-questioned mode driven by AskUserQuestion.
Fetch web content as clean markdown using Cloudflare's markdown.new service. Use this whenever you need to read HTML content from any website - it automatically converts pages to markdown for easier processing.
Create and manage tasks visible in the Claude Code task viewer. Use this skill when you need to track progress on multi-step work, create task lists, or update task status. Essential when the built-in TaskCreate/TaskUpdate tools are not available.
| name | pdf2md |
| description | Convert a PDF file to Markdown. Use this when the user wants to convert a PDF to markdown, extract text from a PDF, or get markdown content from a PDF file. |
| argument-hint | [pdf-path] [output.md] |
| allowed-tools | Bash, Read, Write |
Convert a PDF file to Markdown format.
$0 - Path to the PDF file (required)$1 - Output markdown file path (optional, defaults to same name as PDF with .md extension)First, verify the PDF file exists at the provided path: $0
Determine the output file:
$1 is provided, use it as the output pathEnsure pdf2md is installed. Run this command first (it will skip if already installed):
uv tool install --upgrade ~/.claude/skills/pdf2md/dist/pdf2md-0.1.0-py3-none-any.whl
Run the conversion:
pdf2md "$0" -o "<output-path>"
If successful, inform the user the file was converted and where the output was saved.
If there's an error, explain what went wrong.
/pdf2md document.pdf - Converts to document.md/pdf2md report.pdf output.md - Converts to output.md/pdf2md "My Document.pdf" - Handle paths with spacesIf the user wants specific pages converted, use the -p flag:
pdf2md "$0" -o "<output>" -p "1-5"
If the user wants images extracted too:
pdf2md "$0" -o "<output>" --images