Runnable example
Three-minute AI-coding change
Reproduce a Python text-boundary bug, make a bounded fix, and verify it with the standard library.
The problem
This exercise shows the full evidence chain of a safe AI-coding change: a contract, a failing example, a bounded implementation, and an automated check — in about three minutes.
The contract
- Reproduce the text-boundary bug with a failing check.
- Fix only the starter file with the smallest readable diff.
- Pass the four automated tests, then compare the patch against the reference solution.
Verify it locally
From the root of the FlyPython repository, reproduce the broken starter first, then verify the reviewed solution:
python examples/product-slug/verify.py starter --expect-failurepython examples/product-slug/verify.py solutionHand it to a coding agent
- Clone the repository and reproduce the starter failure with the command above.
- Give the task contract ↗ to your coding agent and allow changes only in
starter/product_slug.py. - Run
python examples/product-slug/verify.py starterand treat any failure as unfinished work. - Compare the patch with the reference solution ↗ — the goal is the same contract with a small, readable diff, not identical syntax.
Keep going
Where this example sits in the roadmap.
Reviewed 2026-09-02 · Repository content version 1. Continue with the related learning path, the repository directory, or the MCP migration guide for the specification background.