Content Search

Content Search lets you search inside document files — not just by filename, but by the actual text content. Find that one PDF with the contract clause, or the Markdown note mentioning a specific topic.

Supported File Types

• PDF — .pdf
• Markdown — .md
• Word — .docx, .doc
• Excel — .xlsx
• PowerPoint — .pptx
• Rich Text & Web — .rtf, .html, .htm
• Plain text & data — .txt, .csv, .tsv, .json, .xml, .yaml, .toml, .log, .srt and more
• Source code — .swift, .py, .js, .ts, .go, .java, .kt, .c, .cpp, .rs, .rb, .php, .sh, .sql, .css and more

Text files in UTF-8 or GB18030 are supported. Files inside node_modules, .git, caches and build output are skipped so large code folders index quickly.

How to Use

1. Select a scope — Choose a directory from the Scope dropdown in the filter toolbar (e.g., Documents, Desktop, or use "Choose Folder" to pick any folder).
2. Enable Content Search — Toggle the Content switch on. It appears in the filter toolbar after the case sensitivity picker. The switch is only available when a scope is selected.
3. Type your query — Enter the text you want to find inside documents. Results will show files that contain matching text.

Search Results

When Content Search is active, results display differently:

• Each result shows the filename and path as usual.
• Below the filename, a matched snippet shows the surrounding text with your search terms highlighted in bold.
• If a file has multiple matches, click "N more matches" to expand and see all snippets.

Indexing

The first time you enable Content Search for a scope, AnySearch extracts and indexes the text content of all supported documents in that directory. This may take a moment depending on the number and size of files.

• A progress bar appears in the status bar showing indexing progress (e.g., "Indexing 42/150").
• You can cancel indexing at any time by clicking the × next to the progress bar.
• Once indexed, subsequent searches in the same scope are instant — the index is cached.
• If a file is modified, its content index is automatically refreshed on the next search.

Natural-language content intent

In 3.0, phrases such as “containing”, “about” or “mentions” can be recognized by natural-language search. When a directory has already been selected and content-indexed, AnySearch searches that local content index first; otherwise it safely falls back to filename search. It never starts a whole-disk content crawl implicitly.

Limits and Performance

Tips

• Content Search works best with text-based documents. Scanned PDFs (images without OCR) won't return results.
• Use a narrow scope for faster indexing — searching your entire home directory will take longer than searching a specific project folder.
• Content Search is a free feature — no Pro subscription required.
• Multi-word queries match files containing all the words (AND logic).