Test whether code comments help your agent

Deleting 168k lines of comments: same scores, ~22% fewer tokens on a small task, three runs per branch.

In short

What it does: Deleting 168k lines of comments: same scores, ~22% fewer tokens on a small task, three runs per branch.

Agent: Claude · Category: Coding

Steps in brief:

  1. Make a copy of the codebase with every comment removed.
  2. Ask the agent to build the same features on both versions, three times each.
  3. Score with lint, type checks, existing tests and a hidden test; compare tokens and cost.

Who made it: openstation.blog (blog, 13 Oct 2026) · source

Last updated · based on 336 use cases on agentsetups, each credited to its author.

by openstation.blog13 Oct 2026

Prompt for your agent

Goal: Measure whether comments in my codebase help or hurt my agent.
Setup: agents used in source: Claude.
Steps:
1. Make a copy of the codebase with every comment removed.
2. Ask the agent to build the same features on both versions, three times each.
3. Score with lint, type checks, existing tests and a hidden test; compare tokens and cost.
Rules: ask me before anything irreversible (sending, posting, paying, deleting); don’t invent data, results or tools I haven’t given you; tell me what you couldn’t verify.
Based on: openstation.blog — https://openstation.blog/2026/10/13/i-deleted-168000-lines-of-comments-my-ai-agent-didnt-notice/

Steps are from the source. Includes a “Based on” credit line.

I tried this

Tell the next builder. One anonymous answer each; click again to undo.

Agents, models and tools only what the source states

Agents
Claude

Sources & credit