AI ZIP File Readers: Inspect an Archive With an Assistant
Most AI assistants cannot read ZIP archives directly. Here is how to check what your tool supports, extract files when needed, and review a collection safely.
Updated 5 min read
Most AI assistants, including Claude and ChatGPT, do not read the contents of a ZIP archive when you upload or attach one. You need to extract the files first and then supply the relevant ones individually. fenbs keeps a map of what you have extracted, what you have reviewed, and what you found, so the next session can continue from where the last one stopped rather than starting the review again from scratch.
The exception is AI coding tools: Claude Code, GitHub Copilot, and similar tools that operate on a file system can often read files extracted from an archive in a project directory. Check your specific tool's documentation before assuming either capability.
Check what your AI tool supports
Before uploading a ZIP, check whether the tool can accept it. Most chat-based AI assistants (Claude.ai, ChatGPT, Gemini) support uploads of common file types — PDFs, images, text files, spreadsheets, code files — but not compressed archives. Uploading a ZIP will usually result in an error or the tool treating the file as a binary it cannot read.
AI coding tools that run against a local directory or a repository can read files from a folder you have already extracted. These tools work on your file system rather than through a file upload, so the ZIP format itself is less of a barrier — the question is whether the extraction step happens before you start the session.
Extract the archive safely
Before extracting an archive from an untrusted source, be aware that ZIP files can contain path-traversal payloads: file paths designed to write to locations outside the target folder. Use a trusted extraction tool rather than double-clicking an unknown archive, and extract to a dedicated folder you can review before moving anything elsewhere.
- On Mac: use the built-in Archive Utility or a tool like The Unarchiver from the App Store.
- On Windows: use the built-in extraction dialog or 7-Zip, which shows the file list before extracting.
- Before extracting, scan the archive with your operating system's built-in security or a dedicated tool if the file came from an external or unknown source.
- Extract to an empty folder so you can see exactly what the archive contained before any files are mixed with your existing work.
Review the file list before uploading
Once extracted, look through the file list before deciding what to upload. Large archives may contain hundreds of files, many of which are not relevant to your question. Uploading only what you need keeps the context focused and reduces the chance of the AI making an error about a file you did not mean to include.
Write a short map of what the archive contained: the main folders, the file types, and the purpose of each major section. This takes a few minutes and is useful both for your own reference and for the next AI session, which can read the map rather than you describing the structure from memory.
Upload and review file by file
Once you have identified the relevant files, upload them one at a time or in small groups. Ask specific questions about each file rather than uploading everything and asking for a summary. "Summarize the contents of this contract" or "List the functions defined in this file" gives a cleaner answer than "Tell me about these files."
For code files, ask the AI to describe what the code does and flag anything that looks unusual, then review the flagged sections yourself. For documents, ask for a summary and then check the original for any claims you plan to act on.
Keep a file map and findings between sessions
In fenbs you can save a context note with the archive's file structure, what you have reviewed so far, and any findings worth noting. When you return to the review in a new session, your connected assistant can read that map and continue from the last reviewed file rather than starting at the beginning. Add a task for each section of the archive that still needs review.
Create a free accountFree for one person · No credit card · Works with Claude, ChatGPT, Cursor and more
How to do this with fenbs
- Create a free account and add a project for the archive review.
- After extracting, add a context note listing the folder structure and file count. Note the file types and the purpose of each major section.
- Create a task for each file group that needs review. Move each task to In Progress when you start it and Completed when you finish.
- Add findings as context notes on the project as you work through each file. Include any decisions about how to handle a particular file type.
- Connect Claude over MCP from Integrations and ask it to read the file map and completed reviews before each session.
When an AI tool can read ZIP files directly
Some AI developer tools and agentic systems can extract and read ZIP archives as part of their tool set. If you are working in a tool that runs code on your behalf, check its documentation for archive support. Claude Code, for example, can be instructed to unzip a file as part of a task when given access to a terminal, and will list the extracted contents before proceeding further if you ask it to confirm what it found.
For context on how AI tools work with files in general, see AI context files compared. For general guidance on how AI assistants handle different file types, check the documentation for the specific tool you are using, as support changes with each update.