What it does
A completion check for coding agents that asks for evidence of edits and tests, not just “done.”
Agent tooling
Canny
Bookmark: Canny
Keep this in your collection.
Leave a noteWhat would you try with this?
: Canny
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
A developer can add this tool to an AI coding assistant. When the AI changes a file and tries to finish, the tool checks its records. If the AI did not run a test after the edit, the tool stops it. The tool then tells the AI to test the work.
The tool needs an access key to connect to TypeSafe. TypeSafe is an outside AI service that reads the assistant's messages. It decides if the assistant is claiming to be done. However, the tool cannot tell if a project's tests actually prove the new code works.