Nano Pdf

• You need to make a quick, targeted text edit to a single PDF page without opening a GUI editor.
• You want to automate PDF corrections in a script or AI agent workflow using natural-language commands.
• You're iterating on slide decks or documents and need to patch specific pages (e.g. update a title, fix a subtitle typo) without re-exporting from the source file.
• You prefer a CLI-first tool that can be chained with other commands in a pipeline.

• You need to edit many pages at once or perform structural changes (reordering, merging, splitting) — nano-pdf targets individual pages with a single instruction.
• You require pixel-precise layout control or need to manipulate images, vector graphics, or embedded fonts; this tool focuses on natural-language text edits.
• You need guaranteed output fidelity without manual review — the SKILL.md explicitly recommends sanity-checking the output before use.

Describe your edit in plain English and nano-pdf interprets and applies it to the target PDF page. No need to write code or learn a PDF manipulation API.

Specify exactly which page to modify as part of the command, keeping edits surgical and avoiding unintended changes to the rest of the document.

Built as a command-line tool, nano-pdf integrates naturally into shell scripts, AI agent pipelines, and automated workflows. Install via uv and invoke it with a single command.

Run nano-pdf edit deck.pdf 1 "Fix the typo in the subtitle" to correct a mistake on the title slide without re-exporting from PowerPoint or Keynote.

Use natural-language instructions to update quarter labels, dates, or names on specific pages of a recurring report — ideal for light automation in an agent workflow.

Catch a last-minute error in a PDF before sending it out? Apply the fix directly from the terminal with a single command, then review the output PDF before distributing.