Chart Kit
Cambi una riga di dati — il grafico si disegna da solo
Snippet di grafici in un unico file, pronti da incollare: linee, barre, torta. Canvas scritto a mano, zero dipendenze — modifichi l'array di dati in JS e hai finito.

Installala in 3 passaggi
Nessuna registrazione. Nessuno strumento da riga di comando. Scegli la tua AI, copia una riga e incollala.
Preferisci fare da te? Acquista la skill ($4.99), scarica lo ZIP dalla conferma d'acquisto, decomprimilo nella cartella mostrata sopra e riavvia la sessione della tua AI — stesso risultato. L'hai già acquistata? Il tuo assistente AI può scaricarla per te.
Prove dal mondo reale
Persone vere hanno svolto proprio questo lavoro con agenti IA.
Pull Substack post stats via computer use
Leggi il caso →Caso d'usoRun company departments with persistent self-hosted agents
Leggi il caso →Caso d'usoRun Monte Carlo simulations on a retirement portfolio
Leggi il caso →Cosa contiene
Un file SKILL.md (la guida operativa) + un modello di partenza. Un'anteprima:
Chart Kit
When to use
- A report or slide deck needs a chart, but you don't want to install ECharts or wrestle with complex config
- A web page needs a simple data graphic: one file, copy-paste, done
- Data changes often and you want "edit the numbers in the array, the chart follows automatically"
- Data for a client or boss must be clean, clearly labeled, and non-misleading
- Offline or intranet environments where external chart CDNs are off-limits
Workflow
- Pick the chart type (input: data + what you want to say → output: chart recommendation)
Line for trends (time series), bar for comparisons (≤8 categories), pie for shares (≤6 slices, with an "Other" catch-all). Nail the one-sentence takeaway first (e.g. "Q3 sales up 40% MoM"), then pick the chart that tells it. - Prep the data (input: raw data → output: JS array)
Reformat into the template's array format: line/bar use{labels:[], values:[]}, pie uses a{name, value}list. Check: units consistent, missing values marked null (the line breaks — never invented), time labels continuous. - Apply the template (input: array → output: working HTML)
Copy the matching chart block from template.html, paste in your data array, update the title and units. Use the default 6-color palette (colorblind-friendly); never use red/green contrast to mean anything other than "bad vs. good". - Add annotations (input: chart → output: annotated chart)
Three required items: ① title (with time range) ② data source (one small line) ③ key-point annotations (e.g. dashed line + label at the peak). Axes must carry units; the Y-axis starts at zero by default. - Anti-misleading check (input: annotated chart → output: pass/fail)
Walk through each: does the Y-axis start at zero (if not, the reason must be stated on the chart), do pie slices add to 100%, are bar widths consistent, is anything using area/3D effects to exaggerate differences. Fix whatever fails. - Deliver (input: checked HTML → output: usage notes)
Tell the user three things: which lines hold the data, how to edit (change the array only), how to reuse (copy the whole chartblock). Include the line: "To change the data, you never touch the code."Output spec
Anteprima del modello
Data Chart Kit · Demo /* Page styles: demo only — each chart block can be copied out on its own */ body{font-family:-apple-system,"PingFang SC","Microsoft YaHei",sans-serif; max-width:720px;margin:0 auto;padding:24px 16px;color:#333;background:#fafafa} h1{font-size:24px;margin-bottom:8px} .chart{background:#fff;border:1px solid #e5e5e5;border-radius:12px; padding:16px;margin:20px 0} .chart h3{margin:0 0 4px;font-size:17px} .src{font-size:12px;color:#999;margin-top:8px} canvas{width:100%;height:260px;display:block;cursor:default} .tip{position:fixed;pointer-events:none;background:#333;color:#fff; font-size:12px;padding:4px 10px;border-radius:6px;display:none;z-index:99} Data Chart Kit
Il pacchetto della skill (SKILL.md + modello) è scritto in inglese.
Potrebbe piacerti anche


