Menu
Generate bar, line, pie, or scatter charts from CSV files or inline data. Pipe table query output into chart.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/k2b-dev/nessi --skill chartThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Build applications with the @k2b/nessi TypeScript library. Use this skill whenever the user wants to create or modify a CLI, backend endpoint, service, prototype, provider switcher, streaming UI adapter, tool-calling workflow, structured-output task, agent loop, or AI integration using @k2b/nessi. Trigger for questions about root nessi(), nessi.structured(), @k2b/nessi/ai complete(), stream(), provider setup, responseFormat, OpenAI/OpenRouter/vLLM/Ollama/Anthropic/Mistral/Gemini, message formats, multimodal input, canonical block streaming events, issue events for malformed tool streams, dynamic per-turn tool resolution, tool call IDs, loopId correlation, loop_end.aggregate metadata, loop-level stats, local or durable steering, historical tool results, context growth, tool execution events, usage accounting, generation options, provider timeouts, API keys, local models, context-overflow handling, or when choosing whether the provider-only /ai layer is enough versus the root APIs.
Developer utilities: Base64, JWT decode, hashing, UUID, passwords, timestamps, regex test, URL encode. Use instead of opening external websites for dev tools.
Search the web and extract page content. Use for current facts, docs, versions, or any URL. Not for github.com — use the github skill.
Based on SOC occupation classification
Showing SKILL.md
| name | chart |
| description | Generate bar, line, pie, or scatter charts from CSV files or inline data. Pipe table query output into chart. |
| metadata | {"nessi":{"command":"chart","enabled":true}} |
Create data visualizations as SVG. Charts render inline in the chat with a download button.
There are 4 chart types: bar, line, pie, scatter. Each reads data from a CSV file or accepts inline values.
| Type | Best for | X axis | Y axis |
|---|---|---|---|
bar | Comparing categories | Category names | One numeric column |
line | Trends over time | Time labels | One or more numeric columns |
pie | Proportions of a whole | Category names | One numeric column |
scatter | Correlation between two numbers | Numeric column | Numeric column |
Always use this when data comes from table query. Pass the CSV path as the first argument:
chart bar /output/data.csv --x "region" --y "revenue" --title "Revenue by Region"
chart pie /output/data.csv --x "category" --y "amount" --title "Distribution"
chart line /output/data.csv --x "quarter" --y "revenue,cost,profit" --title "Trends"
chart scatter /output/data.csv --x "frequency" --y "volume" --label "customer" --title "Segmentation"
--x "column" — the label/category column (required)--y "column" — the value column (required). For line, use comma-separated columns for multiple series--label "column" — (scatter only) show point names next to dots--title "text" — chart title--output /output/name.svg — custom output pathFor quick charts without a file. Use --labels and --values:
chart bar --labels "Jan,Feb,Mar" --values "100,200,150" --title "Sales"
chart pie --labels "Chrome,Firefox,Safari" --values "65,20,15" --title "Browsers"
chart line --labels "Q1,Q2,Q3,Q4" --series '{"Revenue":[100,200,150,250],"Cost":[80,120,100,180]}'
Note: scatter does not support inline mode — it always requires a CSV file.
This is the typical workflow for data analysis:
# Step 1: Aggregate data from a spreadsheet
table query /input/sales.xlsx \
--select "region, sum(revenue) as Revenue, count() as Orders" \
--group "region" --sort "Revenue desc" \
--output /output/by-region.csv
# Step 2: Create chart from the aggregated CSV
chart bar /output/by-region.csv --x "region" --y "Revenue" --title "Revenue by Region"
# Step 3: Display inline with download button
present /output/bar-chart.svg
table query /input/data.xlsx \
--select "customer, count() as Frequency, sum(amount) as Volume" \
--group "customer" --limit 20 \
--output /output/segments.csv
chart scatter /output/segments.csv --x "Frequency" --y "Volume" --label "customer" --title "Customer Segmentation"
present /output/scatter-chart.svg
table query. Do not manually copy values./output/ by default.present after generating a chart to display it inline.