Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesMichael Murphy’s “Time Travel Coding” workflow puts an app’s description in a Markdown file before implementation begins. The idea is to discover unclear requirements and refine the intended experience while changes are still edits to a plan—not code. That may help avoid rework, but Murphy’s article does not measure token savings, so “stop burning tokens” is a rationale, not a proven result.
What “Time Travel Coding” means
In his September 30, 2026, DEV Community article, Michael Murphy argues that a coding agent can spend effort building a version of a program the requester later changes. His alternative is to explore the program in a plain-language Markdown description first, then ask an agent to help imagine and refine the result before implementation. The software remains the eventual outcome; the Markdown file is where the idea gets revised early.
As an Amazon Associate I earn from qualifying purchases.
Murphy summarizes the principle as “Iterate the plan, not the program.” The practical distinction is between changing a written description and asking an agent to revise working code. The approach aims to catch wrong turns before they become implementation work; it does not guarantee that a plan will prevent rework.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteHow to use the workflow
-
Describe the idea in plain language
Start a Markdown file with who the program is for, what it should do, and how it should feel to use. Focus on the intended experience rather than implementation details you have not decided yet.
-
Ask the agent to picture the finished program
Murphy’s sample prompt is: “Can you see what this looks like when it’s finished?” Ask for a screen-by-screen description. Treat the response as a way to expose assumptions and missing details, not as a final specification.
-
Find gaps and update the plan
Ask what is missing, confusing, or could be improved. Choose the useful suggestions and write them into the Markdown file. This keeps decisions in one place instead of leaving important requirements scattered through a conversation.
-
Consider how the idea could grow
Murphy suggests imagining what the program might look like if it continued growing at its current pace for 30 years. This is a prompt for surfacing possible constraints or design pressure, not a forecast and not a requirement to build every imagined feature.
Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy. -
Repeat until suggestions lose value
Continue asking questions and revising the plan while the agent is identifying meaningful gaps. Murphy’s proposed stopping signal is when suggestions become small or repetitive. It is a judgment call, not a measurable threshold.
-
Implement from the revised plan
Once the plan is clear enough to guide the work, ask the agent to build. The point is not to specify every coding decision in advance; it is to settle the intended audience, behavior, and experience before implementation.
Include visual rules the agent should preserve
If appearance matters, write down constraints that should not be broken. Murphy’s examples include avoiding glowing gradients or nested cards, using one accent color, and including the real words on every screen. These are examples from his article, not universal design rules; choose constraints that fit your product.
After implementation, Murphy recommends asking the agent to open the app in a browser, capture a screenshot, and check it against the written rules. That gives the agent a concrete reference for spotting mismatches between the plan and the result. It is a suggested review step, not evidence of a controlled test or a guarantee of visual fidelity.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
What the method does—and does not—establish about token use
Murphy’s case for the method is qualitative: changing a plan is cheaper than rebuilding code, and a fuller plan may help an agent avoid wrong turns. His article reports no token counts, cost comparison, sample size, or controlled productivity results. There is no supported percentage, dollar amount, or token figure for what this workflow saves.
Official guidance supports only a narrower point. Anthropic’s Claude Code guidance recommends considering Plan Mode or asking for a list of files and intended changes before implementation on work affecting multiple files. OpenAI’s Codex guidance says usage depends on the model, where the task runs, task complexity, context, reasoning, speed, and tools. Neither source verifies that Murphy’s Markdown method reduces usage.
That variability matters: even a careful plan does not make every task consume the same amount of usage. The defensible expectation is that planning may reduce avoidable revisions in some projects; whether it lowers token use, and by how much, is not established.
Quick Recap
When planning first is useful
- Try it when: the app’s screens, audience, or behavior are still fuzzy; several screens or files are involved; or visual consistency is important.
- Keep it lightweight when: the change is already clear and limited. A long planning loop can itself take time without improving the result.
- Keep the plan revisable: the goal is to make important decisions visible before building, not to predict every detail or prevent all changes.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
