conon.ai

conon.ai/Tips

Comment The Diff Like A Reviewer

Open the diff panel and attach comments to exact lines, the way reviewers do in real teams — the agent reads them and fixes precisely those places. It beats describing coordinates in chat ('the second paragraph in the services block…') and it spends tokens only on what you flagged.

AdvancedWORKFLOWClaude Code

Why it works

Chat prose approximates locations ('the second paragraph in the services section, no, the other one'); a line comment IS the location. Commenting the diff like a code reviewer aims the agent at exactly the flagged lines — precise, cheap, and identical to how professional teams review, except your reviewee never sulks.

How to do it

  1. Click the +/− change counter to open the diff panel.
  2. Read file by file: left list is what changed, right pane is how.
  3. Hover the exact line you dislike; add a comment right there.
  4. Write the instruction in the comment: 'shorter — one sentence' or 'wrong price, use 49'.
  5. Comment every spot in one pass, then send them together.
  6. The agent fixes precisely the flagged lines — no archaeology, no collateral edits.

Copy this

prompt
I've left comments on specific lines in the diff. Address each comment exactly where it sits — change nothing that isn't flagged.

What to watch for

  • Mixing channels ('see my comments, and also make the header nicer somehow') reintroduces the vagueness the comments just eliminated — keep broad requests as separate prompts.
  • The panel needs a version-control baseline to diff against; no baseline, no panel — one more reason the v1 commit comes first.

Sources

Original: conon.aiDocumentation: code.claude.com

Read next

Scheduled Agents Need An Open Lid
Local scheduled tasks run only while the app is open and the machine is awake — a missed slot is skipped, not postponed. Turn on keep-awake, leave the laptop lid open, and write drift-tolerant instructions: 'if it's already past 5pm, skip today and report that you skipped'. Need lid-closed reliability? That's a cloud task — which can't see your local files.
Claude Code
Typed It Three Times? Make A Skill
The third time you type the same procedure is the signal: say 'create a skill that does this' and the agent writes the skill file itself. Then improve it in a loop — run it, note what's off, have it edit its own skill. House rules: 2nd repeated mistake → CLAUDE.md line; 3rd repeated prompt → skill.
Claude Code
Writer Session, Reviewer Session
Run two parallel sessions on one version-controlled project: one builds, the other — with fresh context — reviews. The docs are explicit about why it works: a fresh session isn't attached to code it didn't write, so it critiques honestly where the author session defends.
Claude Code
Send Noisy Research To A Subagent
Any side-investigation that would flood your session — 'which of these fifty files mentions the old pricing?' — goes to a subagent. It reads everything inside its own separate memory and returns three lines; your session's context stays clean for the actual work.
Claude Code

We build and operate the systems behind tips like this one — agents, automations and internal tools, shipped into production. See what the studio does →