FileConcat

FileConcat merges multiple files, including PDFs and Office documents, into a single file sized for AI chat upload limits. Users drop a folder, files, or a .zip archive, or paste a public repository or page URL. The result screen shows the output file, its token count, the number of files included, and the estimated cost to send it at a chosen model's input price. Documents such as PDF, Word, Excel and slides are converted to text in the browser, and the page states which content was not read in full. The tool also offers a 'Will it fit?' view comparing repository sizes against 128K, 200K and 1M context windows, a browser clipper for YouTube transcripts, Reddit and Hacker News threads, and AI chat conversations, and a command-line version on npm. The page states files are not uploaded to FileConcat, that the tool needs no account, and that anonymous usage analytics run.

Key features

• Combine a folder, files, or .zip/7z/rar/tar archive into one file
• Paste a public GitHub, GitLab, Bitbucket, Gist, or URL to fetch content
• Shows token count, file count, and estimated cost to send
• Converts PDF, Word, Excel, and presentation files to text in the browser
• Reports what content was not read in full, such as images or held-back files
• Filters files by default, with options to override per file or file type
• Context window fit view for public repositories (128K, 200K, 1M)
• Download output as XML, Markdown, or plain text, or copy it
• Browser clipper for YouTube transcripts, Reddit and Hacker News threads, and AI chat conversations
• Command-line version available on npm (@fileconcat/cli)

Best suited for

• Users who hit file upload limits in ChatGPT, Claude, or Gemini
• Combining many documents into one file for AI projects or notebooks
• Preparing codebases or repositories as text for AI chat
• Clipping YouTube transcripts or online discussion threads for AI use

Pricing

Not stated on the website