How to Search Inside AI, PDF, and Sketch Files on Mac

Three of the most common file formats in a designer's library — and macOS gives you almost nothing for searching inside any of them.

Spotlight can find a PDF if you know its name, and sometimes surfaces PDF content if you're lucky with how it indexed the file. But for Illustrator AI files and Sketch files, it reads the filename and stops there. The text inside those files — the headlines, the copy, the artboard names, the symbol names — is invisible to every native macOS search tool.

reverscan extracts full text content from all three formats during normal indexing. Here's exactly what it reads from each one.

Searching Inside AI Files

Illustrator's native format (.ai) is built on PDF. This turns out to be useful: reverscan treats AI files the same way it treats PDFs, passing them through its PDF text extraction pipeline using PDFKit.

What this means in practice: any text that exists as editable text in your Illustrator file — headlines, body copy, labels, artboard names entered as text objects — is fully extracted and indexed. No Deep Scan required. The moment reverscan scans a folder containing AI files, their text content is immediately searchable.

A search for "Thornbury Q3" will surface an AI file containing those words in a text layer, even if the filename is "brand-refresh-FINAL-v2.ai."

[SCREENSHOT: reverscan search results showing an AI file surfaced by a keyword that appears inside the file's text content]

What it extracts: All editable text from text objects in the Illustrator file — including body copy, headlines, labels, and artboard-level annotations. Text embedded inside placed images or converted to outlines is not extracted (those are raster or path data, not text).

Searching Inside PDF Files

PDF content search is reverscan's strongest extraction. It uses Apple's PDFKit to read the full text layer of any PDF — the same framework macOS uses for Preview and the system PDF renderer.

Every word on every page of a text-based PDF is indexed. Search for any term that appears anywhere in the document and reverscan will surface it. This includes multi-page documents, branded templates, specification sheets, client briefs, and any PDF export that preserves text layers.

For PDFs with no text layer — scanned documents, image-heavy exports where text was rasterised — reverscan detects that there's nothing to extract and flags the file for Deep Scan. After a Deep Scan, Apple Vision OCR reads the visible content and indexes it. So even scanned PDFs become searchable, just with an extra step.

[SCREENSHOT: reverscan text search returning a PDF result, showing the filename and path]

What it extracts: Complete text from all pages of text-based PDFs. OCR'd text from image-based PDFs after Deep Scan.

Searching Inside Sketch Files

Sketch files are ZIP archives containing JSON page data. reverscan opens the archive, reads the page files, walks the JSON structure, and extracts the content of every text layer — specifically looking for objects with _class: "text" and pulling their attributedString.string value.

This gets you the actual content of every text layer on every artboard, across every page of the Sketch file. Screen copy, labels, button text, navigation items, placeholder text — all of it indexed without any extra setup.

[SCREENSHOT: reverscan returning a Sketch file in search results for a keyword that appears in a text layer inside the file]

What it extracts: Text content from all text layers across all artboards and pages. Layer names are not separately indexed (unlike PSD), but the text content is far more valuable.

Try it free → Download reverscan — AI, PDF, and Sketch content is indexed automatically, no configuration needed.

Putting It Together: One Search Across All Three

reverscan doesn't search one format at a time. When you type a query, it searches across your entire index — all indexed drives, all supported file types — simultaneously.

A search for a client name or a project phrase will return:

• PDFs containing that phrase anywhere in the document

• AI files with that text in any text layer

• Sketch files with that text on any artboard

All in one results list, ranked by relevance using BM25 scoring with filename matches weighted 10× higher than content matches. The file named "thornbury-brand-guidelines.pdf" will appear above a Sketch file that mentions "Thornbury" in a small label on page 4 — but both surface.

[SCREENSHOT: search results showing a mix of PDF, AI, and Sketch file results for the same query]

Using the Filter Bar to Narrow Results

If a search returns too many results across too many file types, reverscan's filter bar lets you narrow by:

• File type: Images, Documents, or Design files

• Date range: Last 7 days, last 30 days, or last year

• Drive: Filter to a specific indexed drive or folder

Selecting "Design" from the file type chips limits results to AI, PSD, Sketch, Affinity, XD, InDesign, Figma, and IDML files. Combined with a keyword search, this isolates exactly the format and timeframe you're looking for.

What About Other Design Formats?

AI, PDF, and Sketch are the three formats where normal indexing gives you genuinely strong content search. For the rest, Deep Scan unlocks the content.

Frequently Asked Questions

Does this work for AI files saved in older Illustrator formats (e.g., AI8)?
AI files saved with PDF compatibility enabled (the default in modern Illustrator) are fully readable. Very old AI files saved without PDF compatibility may not extract text. If you're not sure, try searching — if the file surfaces on a content query, it worked.

What about Sketch files with multiple pages?
reverscan reads all pages in a Sketch file. Text layers from every page, across all artboards, are indexed.

Can I search for text inside a linked image in a Sketch or AI file?
No. Linked images are raster data — the text in a JPEG placed inside a Sketch artboard is not extractable without OCR. That image file itself can be indexed separately (and OCR'd via Deep Scan), but the text inside placed images within a Sketch or AI file is not captured during normal indexing.

I have thousands of AI and PDF files — how long does indexing take?
reverscan extracts content in parallel using multiple CPU threads. A folder with 1,000 AI or PDF files typically indexes in a few minutes. You can keep working while it runs in the background.

Will new AI, PDF, or Sketch files I save get indexed automatically?
Yes. reverscan's auto-sync (configurable from every 1 minute to once a month) picks up new and modified files on each sync pass. Changes appear in the index without manual intervention.

Does this respect the file structure — can I search within a specific folder?
You can filter search results by drive using the drive selector in the filter bar. Folder-level filtering isn't in the search UI directly, but you can browse the indexed folder tree from the Drives tab to see exactly what's in each location.

Three of the most important file formats in design work, fully searchable by content. Download reverscan free and stop hunting files by name.