Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Turn a one-line objective into a step-by-step construction plan any coding agent can execute cold. Each step has a self-contained context brief — a fresh agent in a new session can pick up any step without reading prior steps.
.claude/skills/sickn33-blueprint/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 708% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -8% | 0% |
Turn a one-line objective into a step-by-step plan any coding agent can execute cold.
Blueprint is for multi-session, multi-agent engineering projects where each step must be independently executable by a fresh agent that has never seen the conversation history. Install it once, invoke it with /blueprint <project> <objective>.
/blueprint myapp "migrate database to PostgreSQL"/blueprint antbot "extract providers into plugins"Do not clone a moving branch directly into an active skills directory. First ask the user to approve network access to the named repository. Then inspect the reviewed revision and ask separately before activating it:
bashreview_dir="$(mktemp -d)" git clone --filter=blob:none https://github.com/antbotlab/blueprint.git "$review_dir/blueprint" git -C "$review_dir/blueprint" checkout --detach 07c5b305cf2d95d584a0d0398c390e839fec5954 git -C "$review_dir/blueprint" ls-files git -C "$review_dir/blueprint" status --short
Read SKILL.md and every bundled file at that exact commit. Check for scripts, hooks, symlinks, network calls, credential access, and instructions that request commands or elevated permissions. Only after explicit user approval, copy the reviewed files into the selected host's skills directory. Re-review a newer revision instead of silently updating this pin.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | fail→fail | 17,916 | 2,279 | -87% | 1 | 1 | 0% | 1,072 | 1,257 | +17% | 0 | 0 | — |
case-01 | fail→fail | 18,537 | 20,332 | +10% | 1 | 1 | 0% | 3,598 | 4,536 | +26% | 0 | 0 | — |
case-02 | fail→pass | 5,353 | 18,033 | +237% | 1 | 1 | 0% | 525 | 4,242 | +708% | 0 | 0 | — |
case-03 | fail→fail | 23,961 | 21,180 | -12% | 1 | 1 | 0% | 4,382 | 4,685 | +7% | 0 | 0 | — |
case-04 | pass→pass | 5,559 | 4,458 | -20% | 1 | 1 | 0% | 923 | 1,690 | +83% | 0 | 0 | — |
case-05 | pass→pass | 3,599 | 3,724 | +3% | 1 | 1 | 0% | 632 | 1,592 | +152% | 0 | 0 | — |
case-06 | pass→pass | 7,959 | 5,548 | -30% | 1 | 1 | 0% | 1,701 | 1,986 | +17% | 0 | 0 | — |
case-07 | pass→pass | 18,555 | 10,502 | -43% | 1 | 1 | 0% | 3,279 | 2,675 | -18% | 0 | 0 | — |
case-08 | pass→pass | 19,046 | 1,390 | -93% | 1 | 1 | 0% | 1,794 | 1,167 | -35% | 0 | 0 | — |
case-09 | fail→pass | 9,715 | 1,729 | -82% | 1 | 1 | 0% | 1,806 | 1,155 | -36% | 0 | 0 | — |
case-10 | fail→pass | 11,837 | 3,646 | -69% | 1 | 1 | 0% | 2,098 | 1,498 | -29% | 0 | 0 | — |
case-11 | pass→pass | 8,944 | 5,916 | -34% | 1 | 1 | 0% | 1,542 | 1,838 | +19% | 0 | 0 | — |
case-12 | fail→pass | 6,135 | 1,559 | -75% | 1 | 1 | 0% | 1,039 | 1,202 | +16% | 0 | 0 | — |
case-14 | fail→fail | 4,797 | 2,365 | -51% | 1 | 1 | 0% | 680 | 1,264 | +86% | 0 | 0 | — |
case-15 | pass→pass | 10,768 | 2,761 | -74% | 1 | 1 | 0% | 1,809 | 1,415 | -22% | 0 | 0 | — |
case-16 | pass→pass | 12,955 | 10,401 | -20% | 1 | 1 | 0% | 2,254 | 2,623 | +16% | 0 | 0 | — |
case-17 | pass→fail | 6,750 | 2,845 | -58% | 1 | 1 | 0% | 1,083 | 1,375 | +27% | 0 | 0 | — |
case-18 | fail→pass | 12,919 | 7,277 | -44% | 1 | 1 | 0% | 2,327 | 2,149 | -8% | 0 | 0 | — |
case-19 | pass→pass | 10,547 | 7,958 | -25% | 1 | 1 | 0% | 1,886 | 2,140 | +13% | 0 | 0 | — |
case-20 | pass→pass | 10,250 | 6,945 | -32% | 1 | 1 | 0% | 1,977 | 2,031 | +3% | 0 | 0 | — |
case-21 | pass→pass | 15,028 | 8,012 | -47% | 1 | 1 | 0% | 2,512 | 2,258 | -10% | 0 | 0 | — |
case-22 | pass→pass | 10,349 | 6,411 | -38% | 1 | 1 | 0% | 1,646 | 1,922 | +17% | 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. 22 cases were attempted, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 20 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.
Other measured skills in the registry, with their headline benchmark lift.