Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Designs a realistic, evidence-based plan to build a new habit or break a bad one — using cue-routine-reward loops, habit stacking, environment design, tiny starting steps, and relapse-proof tracking. Use this skill when a user asks to "help me build a habit", "stop a bad habit", "be more consistent", "create a morning/evening routine", "stick to X", or wants a system to make a behavior automatic.
.claude/skills/jayrha-habit-builder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 33% | 0% |
| case-03 | ✓→✓ | = Same ✓ | -7% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 7% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 36% | 0% |
This skill turns a vague intention ("I want to read more / stop doomscrolling") into a concrete behavior-change plan grounded in habit science — the cue→routine→reward loop, making good habits obvious/easy/attractive/satisfying and bad ones the opposite, habit stacking, environment design, and a tracking system that survives slip-ups.
Keywords: habit, routine, behavior change, consistency, habit stacking, atomic habits, morning routine, break a habit, accountability, streak, willpower, motivation.
Use this for everyday behavior change (exercise, reading, sleep, focus, screen time, diet habits). This is coaching, not clinical treatment — for addiction, self-harm, eating disorders, or compulsions causing real harm, recommend a qualified professional rather than a habit tracker.
references/habit-loop.md.references/environment-design.md.references/tracking.md.| Symptom | Fix | | --- | --- | | "I keep forgetting" | Weak cue → habit-stack onto a rock-solid existing routine | | "I don't have time" | Habit too big → shrink to a 2-minute version | | "I lose motivation" | Relying on willpower → redesign environment + add immediate reward | | "I start then quit" | No recovery plan → adopt never-miss-twice + if-then plans | | "I do it but it doesn't stick" | No identity link → frame as "becoming the kind of person who…" |
See examples/morning-routine.md for a full plan to build a consistent morning routine, including the obstacle plan.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 15,257 | 31,657 | +107% | 1 | 1 | 0% | 2,535 | 3,570 | +41% | 0 | 0 | — |
case-02 | pass→pass | 9,697 | 6,326 | -35% | 1 | 1 | 0% | 1,613 | 2,138 | +33% | 0 | 0 | — |
case-03 | pass→pass | 16,297 | 7,641 | -53% | 1 | 1 | 0% | 2,760 | 2,572 | -7% | 0 | 0 | — |
case-04 | pass→pass | 13,003 | 7,331 | -44% | 1 | 1 | 0% | 2,233 | 2,388 | +7% | 0 | 0 | — |
case-05 | fail→pass | 11,414 | 11,869 | +4% | 1 | 1 | 0% | 1,935 | 2,997 | +55% | 0 | 0 | — |
case-06 | pass→pass | 9,756 | 6,753 | -31% | 1 | 1 | 0% | 1,607 | 2,191 | +36% | 0 | 0 | — |
case-07 | pass→pass | 10,950 | 10,494 | -4% | 1 | 1 | 0% | 1,835 | 2,726 | +49% | 0 | 0 | — |
case-08 | pass→pass | 10,212 | 8,837 | -13% | 1 | 1 | 0% | 1,772 | 2,432 | +37% | 0 | 0 | — |
case-09 | pass→pass | 9,372 | 16,347 | +74% | 1 | 1 | 0% | 1,510 | 2,006 | +33% | 0 | 0 | — |
case-10 | pass→pass | 9,865 | 8,359 | -15% | 1 | 1 | 0% | 1,750 | 2,473 | +41% | 0 | 0 | — |
case-11 | pass→pass | 10,587 | 8,697 | -18% | 1 | 1 | 0% | 1,743 | 2,465 | +41% | 0 | 0 | — |
case-12 | pass→pass | 11,304 | 9,417 | -17% | 1 | 1 | 0% | 1,795 | 2,673 | +49% | 0 | 0 | — |
case-13 | pass→pass | 11,575 | 11,212 | -3% | 1 | 1 | 0% | 1,949 | 2,834 | +45% | 0 | 0 | — |
case-14 | pass→pass | 6,884 | 5,263 | -24% | 1 | 1 | 0% | 1,145 | 1,945 | +70% | 0 | 0 | — |
case-15 | pass→pass | 10,998 | 21,337 | +94% | 1 | 1 | 0% | 1,699 | 2,585 | +52% | 0 | 0 | — |
case-16 | pass→pass | 12,248 | 9,115 | -26% | 1 | 1 | 0% | 2,004 | 2,403 | +20% | 0 | 0 | — |
case-17 | pass→pass | 13,929 | 16,869 | +21% | 1 | 1 | 0% | 2,282 | 2,840 | +24% | 0 | 0 | — |
case-18 | pass→pass | 9,717 | 6,632 | -32% | 1 | 1 | 0% | 1,606 | 2,041 | +27% | 0 | 0 | — |
case-19 | pass→pass | 10,121 | 8,688 | -14% | 1 | 1 | 0% | 1,551 | 2,388 | +54% | 0 | 0 | — |
case-20 | fail→fail | 11,805 | 10,809 | -8% | 1 | 1 | 0% | 1,969 | 2,875 | +46% | 0 | 0 | — |
case-21 | pass→pass | 8,917 | 7,286 | -18% | 1 | 1 | 0% | 1,422 | 2,187 | +54% | 0 | 0 | — |
case-22 | pass→pass | 9,840 | 8,252 | -16% | 1 | 1 | 0% | 1,647 | 2,477 | +50% | 0 | 0 | — |
case-23 | pass→pass | 6,269 | 6,229 | -1% | 1 | 1 | 0% | 1,090 | 2,048 | +88% | 0 | 0 | — |
case-24 | pass→pass | 12,369 | 11,479 | -7% | 1 | 1 | 0% | 1,995 | 2,901 | +45% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +4 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/3/2026 | +14% |
Other measured skills in the registry, with their headline benchmark lift.