Framer AI AgentsDesign and publish professional sites with AI
MashuPack
Turn codebases into a clean file for Claude and ChatGPT
123 followers
Turn codebases into a clean file for Claude and ChatGPT
123 followers
Select the exact parts of a repository and compile them into one clean text file for ChatGPT and Claude or Gemini. MashuPack is built for browser-based AI workflows where file-count limits, upload friction, and messy context assembly keep getting in the way. It makes code context portable, intentional, and easy to control.
Mailwarm
Do you keep file paths and a table of contents so the model can still reference where things came from?
@othman_katim Yes - MashuPack preserves file paths and structure, and the tree/report can be copied along with the full contents. It also already shows token estimates in the UI for folders/files when you toggle the stats view. What I want to improve next is making more of that metadata clear in the exported manifest itself, so the output stays self-describing outside the app. Thank you for your comment!
Great idea, takes me back to the "early" days of chatgpt (v3) before coding tools, constantly copying files back and forth into the browser. Big win for those using the browser still
@willsmithte Thanks so much! I use Claude Code CLI and Codex CLI for code execution, but I still do a lot of long-form planning, debugging, and review in conversational web UIs like ChatGPT and Claude, which is the primary reason for needing this. The parsed headers in the output make it so AI on the web can extract the full file contents into its internal environment and grep/sed around just like in Codex or Claude Code without filling the context window.
Private Resume Builder
nice and simple utility, I've been using tree for this, but Mashupack is a nice visual way to solve the same problem.
A context manifest would be valuable here.
Something like file list, timestamp, ignored paths, token estimate, and exclusions.
That gives teams a lightweight audit trail for what was sent into ChatGPT or Claude.
@nithin_raju1 Thanks so much!