What is Effort Pilot?

Spend 20–50% fewer tokens and get responses ~1.5× faster in Claude Code.

Effort Pilot switches the reasoning effort for every turn, automatically: plans get xhigh, building gets medium, questions and trivial tasks — a git commit, a push, a rename — get low. You stop paying for deep thinking where none is needed, and get more of it where it counts.

Effort Pilot switching effort: xhigh to plan, high to investigate, medium to build, low to commit and push

What you get

Same work, compared with leaving effort on high all the time: fewer tokens burned, faster responses, and fewer turns
you have to redo, because hard work gets more thinking, not less.

Effort Pilot vs. always-on high: about 35% fewer tokens, 1.5x faster responses, 20% fewer turns that fall short

_{≈ Approximate, from the author's own sessions; savings range from ~20% on hard work to ~50% and more on everyday turns. Every setup differs: measure yours with /effort-pilot report.}

Install

In Claude Code:

/plugin marketplace add vadympetryshyn/effort-pilot
/plugin install effort-pilot@effort-pilot
/reload-plugins

Or from your terminal, in one line (then start a new session):

claude plugin install effort-pilot --marketplace vadympetryshyn/effort-pilot

It is on as soon as it loads: from your next message, the line above the prompt shows ⏵ effort with the effort it
picked and why. Nothing to configure — every setting has a default (the terminal install still says its options are
"not yet set": Claude Code counts any option you have not saved yourself, and the defaults apply). Works on Opus and Fable (Sonnet and Haiku are left to you). Needs a recent
Claude Code (tested on 2.1.318).

Where it stands aside. On the newer models (Opus 5 and 5.5, Fable 5.1) with an Anthropic login, effort is set per
message, and a switch costs no prompt cache. Elsewhere effort is one setting of the whole request, and every switch
would rewrite the conversation's prompt cache: on Claude 4.x and older, Fable 5, sessions on Bedrock, Vertex, Foundry or
a gateway, and an effort set as a number. There Effort Pilot leaves your own effort alone (no classifier call either),
and the band says so. On a model it does not know, it watches the cache: if a switch is seen to rewrite it, it stands
aside on that model for the rest of the session.

Update: claude plugin marketplace update effort-pilot && claude plugin update effort-pilot@effort-pilot, then
restart. Pause: /effort-pilot off. Remove: /plugin uninstall effort-pilot@effort-pilot.

How it works

Before each turn a small, fast model (Haiku, ≈ $0.0002 a turn) reads your message and picks the effort:

You…	Phase	Effort
ask for a plan, or use plan mode	plan	`xhigh`
ask a hard question, investigate a bug	analyze	`high`
ask to build or fix something	implement / fix	`medium` (`high` when hard)
approve a plan	implement	`medium` — the thinking went into the plan
ask a quick question, commit, push, run a command, a tiny edit	chat / trivial	`low`

A band above the prompt shows the effort now, how it got there and why. Follow-ups that push the work on keep its effort, and a side question doesn't drag it down; a plain question about the work (what was found, where it stands) runs at low — the answer is already there.

Features

🗣️ Just say it. "do this on max", "medium is enough", "plan it, then build on high" — in any language, typed or dictated. "back to auto" hands it back.
🧭 Plan mode runs at xhigh; the approved plan is built at medium.
🤖 Subagents get their own effort (never below the build effort, medium by default).
🔔 Background results continue at the effort of the work they serve.
✋ You stay in charge. /effort <level> pins it; /effort-pilot on resumes.
🔒 Private. Nothing leaves your machine but the Haiku call. Your effort setting is changed per message; the model is never changed, and the prompt cache stays intact (where a switch would rewrite it, Effort Pilot stands aside).

Commands

Command
`/effort-pilot`	current state
`/effort-pilot on` · `off` · `shadow`	run · stop · only show what it would pick
`/effort-pilot stats`	this session's turns and efforts
`/effort-pilot classify <text>`	what it would pick for a message
`/effort-pilot report`	cost, speed and quality vs. `high` in a pane with charts beside the chat (as text under `claude -p`; needs measure)
`/effort-pilot reset`	delete the turn records measure kept

Settings

All in /config → effort-pilot.

Setting	Default
`planEffort`	`xhigh`	plans and plan mode
`analyzeEffort`	`high`	hard questions, investigations
`buildEffort`	`medium`	building and fixing
`hardBuildEffort`	`high`	building that is hard (many files, tricky logic)
`talkEffort`	`low`	questions, chat, tiny edits
`managedModels`	`opus,fable`	models whose effort it sets (never where effort is one setting per request: see above)
`headless`	off	also steer `claude -p` and SDK runs
`measure`	off	run the comparison below
`controlShare`	`30`	% of turns left at `high` while measuring

Measure it yourself

Turn on measure. A share of turns where Effort Pilot would pick something other than high is left at high (control turns). /effort-pilot report then compares, per phase, cost per turn, model time and turns that fell short (aborted, or your next message says it was wrong), each with a 90% range, and nets out what the classifier itself cost. Records (turn statistics, no prompt text) live in the plugin's data folder, ~/.claude/plugins/data/effort-pilot-effort-pilot/turns/, one file a day, and are kept for 90 days. /effort-pilot reset deletes them; uninstalling the plugin removes them too (unless you uninstall with --keep-data).

Contributing

Issues and PRs welcome — see CONTRIBUTING.md. Questions and ideas go to
Discussions; a turn that got the wrong effort is the most
useful issue.

If it saves you tokens, a ⭐ helps others find it.

License

MIT