Two ready-to-use prompts to make coding agents fix happy-path debt and implicit state machines
Yamapama · x · 2026-11-15
X user Yamapama shared two copy-paste prompts for coding agents (Claude, Codex, etc.):
- Happy-path debt scan: have the agent find every screen that only works when data loads fast, nothing fails, lists aren't empty, text is short, and the screen is wide — then fix missing states and layout breaks one component at a time, without changing the design, and summarize before/after by file
- Implicit state machines: find all implicit state machines in the app, plan first without touching code, optimizing UX, DX and agent experience (AX) together
A reply notes this kind of "clean up and make more efficient" task works particularly well in Claude. Both prompts hinge on the constraints "plan first, change one thing at a time, break nothing".
More from coding & agent
- Explainer animation built entirely with code, no video editor needed — iamfakhrealam · 2026-11-15
- Alibaba's Qoder quietly adds stealth Qwen4-Max-Preview — one-prompt 3D Earth in 10 minutes — iamfakhrealam · 2026-11-15
- lattedit: free local-first browser LaTeX editor with agent API takes on Overleaf — cneuralnetwork · 2026-11-15
- Garry Tan: Only ~1% of engineers let agents ship features — it should grow 100x — garrytan · 2026-11-15
- ArcGrav: A Prompt-Native Agent Control Language to Cut Instruction Overhead — ArchitecturalGravity · 2026-11-15
- Prompting regime is back: scaffolding and context now separate vastly different results from same models — Afinetheorem · 2026-11-15