AI Badger
Local-first tool that prepares focused repo context to paste into any AI chat for review.
Tested on Linux
Use it
Linux x86_64
Download install script
curl -fsSL https://raw.githubusercontent.com/PVRLabs/aibadger/main/install.sh -o /tmp/install.shInstall badger via script
sh /tmp/install.sh
Check that it works
badger --helpThe output should contain “AIBADGER - local repository context and review bridge”.
First thing to try
mkdir -p /tmp/badger-demo && printf 'def add(a, b):\n return a + b\n' > /tmp/badger-demo/calc.py && printf '# Demo\nA tiny project for badger smoke test.\n' > /tmp/badger-demo/README.md && printf 'print("hello")\n' > /tmp/badger-demo/main.py && ls -R /tmp/badger-demo && echo SETUP_OKcd /tmp/badger-demo && git init -q && git add . && git -c user.email=t@t.t -c user.name=t commit -qm init && echo "second line" >> calc.py && git diff --stat && echo GIT_OKbadger map /tmp/badger-demoRun badger map <repo> in a repository to generate an offline HTML map of the code.
What it does
- Get a second opinion on code changesTested
How to use it
- Make your changes in a git repository.
- Run
badger api review-context --root <repo> --mode defaultto build a review prompt with your changes. - Paste the output into any AI chat.
- Extract focused repo context for AI chatTested
How to use it
- Write your goal in a text file, and list the files you want in a selectors file, one per line, such as
FILE:<file>. - Run
badger api extract --root <repo> --focus code --input <selectors-file> --goal-file <goal-file>. - Paste the output into any AI chat.
- Write your goal in a text file, and list the files you want in a selectors file, one per line, such as
- Generate an interactive offline codebase mapTested
How to use it
- Run
badger map <repo>to create an HTML report of the codebase. - Open the generated HTML file in a browser.
- Run
- Review and apply AI-suggested changesNot shown working
How far pickfit got
Linux x86_64
- Installs
- Starts
- 3 of 4 features work
- Works offline
Good for
- get a second opinion on code changes via AI chat
- generate an offline interactive HTML map of a repository
- run focused review or design workflows on a repository
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
Its main function works
- Replayed from scratch
- Works offline
- Checked on 30 Oct 2026
- Version v0.8.3
Differences the test machine hit
- README: badger, badger review
In the test: use stable non-interactive API (badger api review-context / extract / review-continuation) with --root and input files for headless verification; interactive badger / badger review need a TTY with clipboard/browser - README: badger map to generate an interactive, offline HTML map
In the test: badger map /tmp/badger-demo wrote HTML to cwd (/home/sbx/work/badger-codemap-badger-demo.html), not inside the target dir