~/resources/skills/realistic-eta
// skill

Realistic-ETA

Stops Claude from saying "should be quick" when something will actually take an hour. Six rules for calibrated time estimates, applied any time you're tempted to write the words quick, small, simple, easy, or just.

// what it does

The number-one way Claude burns trust with operators is by saying "should take 5 minutes" when the actual delivery is 45. The optimism feels generous in the moment. It isn't. It blows your plan for the next hour, makes every subsequent commitment feel cheap, and accumulates into a pattern of un-reliability that's hard to climb out of.

Realistic-ETA is the calibration layer. Six rules. Take your honest first guess and double it. Distinguish code-time from verify-time — writing the change is half the work; making sure it's correct is the other half. Reject vague qualifiers like quick, small, simple, easy, just — replace each with a number even if you have to invent it. Name what could blow up before locking the estimate. Round up, never down. And if the task really is 2 minutes, say "2 minutes" — the rule isn't always-inflate, it's stand-by-a-number-you-can-be-measured-against.

The most useful trigger isn't a slash command. It's whenever Claude is mid-sentence about to write "should be quick." That's the moment the skill fires — stop, run the rules, replace with a number.

// how to install
  1. Download the file below — it's a single SKILL.md.
  2. Save it to ~/.claude/skills/realistic-eta/SKILL.md on your machine. Create the folder if it doesn't exist.
  3. Restart Claude Code so the skill is picked up.
  4. Say /realistic-eta at the start of any session, or just trust it to fire when you ask "how long will this really take?"
trigger:/realistic-eta·/eta
~ download SKILL.md →
// peek inside — rule 3, reject vague qualifiers 
Quick, small, simple, easy, just — these are red flags
in your own output. If you're tempted to use one, replace
it with a specific number even if you have to invent it.

Numbers force commitment; vague qualifiers hide overconfidence.

Output format when asked for an ETA:

  Estimate: ~60 minutes total.
  - Code: 20 min
  - Verify: 20 min
  - Buffer for risk: 20 min

  Risks that could double this:
  - [specific risk 1]
  - [specific risk 2]