Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Find, evaluate, and download low-level common standard CAD parts from step.parts, such as screws, bolts, nuts, washers, bearings, standoffs, electronics parts, motors, connectors, and other off-the-shelf components. Use when Codex needs to search the hosted step.parts catalog, resolve fuzzy part names, standards, aliases, or dimensions, choose a matching part, fetch a canonical .step file, verify checksums, or use the step.parts API/OpenAPI/catalog endpoints for standard part discovery.
.claude/skills/earthtojake-step-parts/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 0% | 0% |
Use the hosted step.parts machine endpoints instead of scraping HTML or relying on local repository files. Treat https://api.step.parts as the canonical API origin and https://www.step.parts as the site/static-asset origin unless the user provides a different hosted mirror. If the domain does not resolve or the API is unavailable, report that the hosted service is not reachable yet instead of falling back to repo-specific assumptions.
q for fuzzy tokens, standards, aliases, dimensions, source/product URLs, and attribute names/values.category, family, standard, or tag when the user gives an exact facet./v1/parts and inspect items, total, and facets.id, name, standard, and key attributes before choosing. If one result clearly matches, return the selected record details without downloading unless the user asked for a local STEP file.stepUrl, then verify the file with the record's sha256 when present.Use scripts/download_step_part.py for deterministic search, download, and checksum verification:
bashpython skills/step-parts/scripts/download_step_part.py "M3 socket head 12" --download --out-dir /tmp/step-parts python skills/step-parts/scripts/download_step_part.py --id iso4762_socket_head_cap_screw_m3x12 --download --out-dir /tmp/step-parts python skills/step-parts/scripts/download_step_part.py "bearing 608zz" --limit 5
Useful options:
--origin: override https://api.step.parts only when the user provides another hosted API origin.--tag, --category, --family, --standard: repeatable facet filters.--out-dir: directory for downloaded STEP files. Defaults to /tmp/step-parts.--all: with --download, download every result on the returned page as individual STEP downloads.--overwrite: replace an existing output file.The script prints JSON to stdout. For searches, it prints matched records. For downloads, it prints saved file paths, checksums, and source URLs.
Read references/step-parts-api.md when you need endpoint details, field meanings, or query semantics. Prefer:
/v1/parts for filtered search with absolute asset URLs./v1/parts/{id} for one enriched record.stepUrl for STEP downloads./v1/catalog/parts.index.json for a compact discovery index./v1/catalog/schema for field and family attribute meanings./v1/openapi.json when generating a client or tool.M3 SHCS 12 before adding exact family and standard filters.tag, category, family, and standard fields are ANDed together. Use exact facets to narrow within known categories, then rank manually by name and attributes.ISO 4762, ISO4762, or the exact standard.designation.attributes object contains family-specific facts such as thread, lengthMm, bore1Mm, material, profileSeries, slotSizeMm, and dimensions in millimeters.https://www.step.parts; STEP URLs are environment-aware and may resolve to GitHub LFS media in production. Use catalog/API stepUrl for downloads.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | pass→pass | 11,955 | 7,402 | -38% | 1 | 1 | 0% | 2,106 | 2,331 | +11% | 0 | 0 | — |
case-01 | fail→fail | 6,658 | 6,720 | +1% | 1 | 1 | 0% | 1,060 | 1,223 | +15% | 0 | 0 | — |
case-02 | fail→fail | 37,847 | 9,158 | -76% | 1 | 1 | 0% | 2,110 | 1,240 | -41% | 0 | 0 | — |
case-03 | fail→fail | 7,130 | 6,022 | -16% | 1 | 1 | 0% | 1,272 | 1,235 | -3% | 0 | 0 | — |
case-04 | fail→pass | 22,093 | 3,691 | -83% | 1 | 1 | 0% | 1,369 | 1,600 | +17% | 0 | 0 | — |
case-05 | fail→pass | 8,461 | 3,227 | -62% | 1 | 1 | 0% | 1,495 | 1,528 | +2% | 0 | 0 | — |
case-06 | fail→pass | 15,398 | 1,955 | -87% | 1 | 1 | 0% | 2,353 | 1,234 | -48% | 0 | 0 | — |
case-07 | fail→pass | 9,786 | 2,182 | -78% | 1 | 1 | 0% | 1,459 | 1,302 | -11% | 0 | 0 | — |
case-08 | pass→pass | 12,385 | 3,674 | -70% | 1 | 1 | 0% | 1,949 | 1,560 | -20% | 0 | 0 | — |
case-09 | fail→pass | 10,401 | 3,771 | -64% | 1 | 1 | 0% | 1,625 | 1,619 | -0% | 0 | 0 | — |
case-10 | pass→pass | 10,699 | 3,220 | -70% | 1 | 1 | 0% | 1,590 | 1,478 | -7% | 0 | 0 | — |
case-11 | pass→pass | 5,701 | 1,686 | -70% | 1 | 1 | 0% | 911 | 1,167 | +28% | 0 | 0 | — |
case-12 | pass→pass | 6,264 | 2,878 | -54% | 1 | 1 | 0% | 1,085 | 1,402 | +29% | 0 | 0 | — |
case-13 | pass→pass | 7,863 | 2,556 | -67% | 1 | 1 | 0% | 1,246 | 1,240 | -0% | 0 | 0 | — |
case-14 | pass→pass | 11,059 | 4,975 | -55% | 1 | 1 | 0% | 1,771 | 1,868 | +5% | 0 | 0 | — |
case-15 | pass→pass | 10,437 | 3,613 | -65% | 1 | 1 | 0% | 1,691 | 1,607 | -5% | 0 | 0 | — |
case-16 | fail→pass | 9,227 | 2,788 | -70% | 1 | 1 | 0% | 1,468 | 1,334 | -9% | 0 | 0 | — |
case-17 | fail→pass | 5,776 | 2,174 | -62% | 1 | 1 | 0% | 807 | 1,221 | +51% | 0 | 0 | — |
case-19 | pass→pass | 13,706 | 1,863 | -86% | 1 | 1 | 0% | 2,241 | 1,222 | -45% | 0 | 0 | — |
case-20 | pass→pass | 35,361 | 29,382 | -17% | 1 | 1 | 0% | 6,168 | 6,800 | +10% | 0 | 0 | — |
case-21 | pass→pass | 10,276 | 11,706 | +14% | 1 | 1 | 0% | 1,837 | 2,909 | +58% | 0 | 0 | — |
case-22 | pass→pass | 10,490 | 9,720 | -7% | 1 | 1 | 0% | 1,952 | 2,597 | +33% | 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 19 counted toward the lift figure. The other 3 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 +32 percentage points is the difference between those two pass rates over the 19 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.