Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Fetch a stock quote snapshot with keyless curl — Yahoo Finance's public chart endpoint, read with the discipline unofficial market data demands: timestamped, delayed-flagged, and never advice. Use when asked what's this stock at, how did the market do today, get me a ticker's recent range, or pull basic price history. Produces the quote with change and range context, the source-honesty caveats (unofficial, possibly delayed), the rerunnable command, and a hard no-advice line.
.claude/skills/mohitagw15856-stock-snapshot/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 826% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 427% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 28% | 0% |
Stock prices are the live-data request with the highest stakes-to-availability mismatch: everyone asks, keyless official sources barely exist, and the standard keyless answer — Yahoo Finance's public chart endpoint — is unofficial, reshapeable, and possibly delayed. This skill uses it the only defensible way: clearly sourced, timestamped, delay-flagged, wrapped in the no-advice line, and honest that anything portfolio-grade belongs with a real brokerage feed. Within those fences, it's genuinely useful: quotes, day moves, and recent ranges in one curl.
Ask for these if not provided:
SAP.DE vs SAP) and the wrong-listing answer is confidently wrong in the wrong currencycurl -s -A "Mozilla/5.0" "https://query1.finance.yahoo.com/v8/finance/chart/AAPL?range=1d&interval=1d" — the meta object carries regularMarketPrice, previousClose, regularMarketTime (epoch — convert and show), currency, exchangeName. History: range=1mo&interval=1d → the timestamp + indicators.quote arrays. The User-Agent header is required — without it the endpoint often refuses.regularMarketTime gets converted and displayed; outside market hours the snapshot says "as of Friday's close" not "currently"; data may be delayed 15+ minutes depending on exchange — flagged, not assumed away.currency and exchangeName fields are printed with the price — a right number in an unstated currency is a wrong number.Price] · ±change, ±%] on the day · as of converted timestamp] (market open / after close])
Day range · volume · on request: 52-week position, the 1-month shape in one descriptive sentence]
Source: Yahoo Finance public endpoint (unofficial; may be delayed) · rerun: [exact curl] Informational snapshot, not investment advice or trading-grade data — decisions belong on a brokerage feed.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 10,131 | 12,644 | +25% | 1 | 1 | 0% | 1,879 | 1,484 | -21% | 0 | 0 | — |
case-02 | fail→fail | 11,146 | 25,870 | +132% | 1 | 1 | 0% | 1,715 | 2,977 | +74% | 0 | 0 | — |
case-03 | fail→pass | 15,903 | 9,479 | -40% | 1 | 1 | 0% | 268 | 2,482 | +826% | 0 | 0 | — |
case-04 | fail→fail | 15,497 | 16,815 | +9% | 1 | 1 | 0% | 223 | 3,330 | +1393% | 0 | 0 | — |
case-05 | fail→pass | 11,960 | 9,490 | -21% | 1 | 1 | 0% | 1,198 | 1,859 | +55% | 0 | 0 | — |
case-06 | fail→pass | 4,292 | 18,790 | +338% | 1 | 1 | 0% | 671 | 3,538 | +427% | 0 | 0 | — |
case-07 | fail→pass | 20,154 | 14,388 | -29% | 1 | 1 | 0% | 2,239 | 2,547 | +14% | 0 | 0 | — |
case-08 | fail→fail | 11,074 | 18,574 | +68% | 1 | 1 | 0% | 1,057 | 3,452 | +227% | 0 | 0 | — |
case-09 | fail→fail | 24,268 | 9,582 | -61% | 1 | 1 | 0% | 2,646 | 1,709 | -35% | 0 | 0 | — |
case-10 | fail→pass | 13,469 | 9,067 | -33% | 1 | 1 | 0% | 1,288 | 1,651 | +28% | 0 | 0 | — |
case-11 | fail→fail | 17,163 | 20,396 | +19% | 1 | 1 | 0% | 1,732 | 3,867 | +123% | 0 | 0 | — |
case-12 | fail→fail | 28,069 | 16,799 | -40% | 1 | 1 | 0% | 1,821 | 3,160 | +74% | 0 | 0 | — |
case-13 | fail→pass | 2,971 | 5,311 | +79% | 1 | 1 | 0% | 426 | 1,962 | +361% | 0 | 0 | — |
case-14 | fail→fail | 21,207 | 9,228 | -56% | 1 | 1 | 0% | 2,292 | 2,649 | +16% | 0 | 0 | — |
case-15 | pass→pass | 16,284 | 15,086 | -7% | 1 | 1 | 0% | 1,836 | 3,111 | +69% | 0 | 0 | — |
case-16 | fail→pass | 11,479 | 9,830 | -14% | 1 | 1 | 0% | 933 | 2,732 | +193% | 0 | 0 | — |
case-17 | fail→fail | 13,293 | 10,136 | -24% | 1 | 1 | 0% | 1,467 | 1,995 | +36% | 0 | 0 | — |
case-18 | pass→fail | 10,513 | 15,408 | +47% | 1 | 1 | 0% | 803 | 2,776 | +246% | 0 | 0 | — |
case-19 | pass→pass | 12,071 | 13,679 | +13% | 1 | 1 | 0% | 1,200 | 1,829 | +52% | 0 | 0 | — |
case-20 | fail→fail | 23,908 | 29,856 | +25% | 1 | 1 | 0% | 4,332 | 5,137 | +19% | 0 | 0 | — |
case-21 | fail→fail | 51,732 | 51,963 | +0% | 1 | 1 | 0% | 8,218 | 8,358 | +2% | 0 | 0 | — |
case-22 | fail→pass | 15,716 | 14,932 | -5% | 1 | 1 | 0% | 2,961 | 3,706 | +25% | 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.