AVA skill
AVA is a visual-analytics skill that answers questions about a CSV or database with inspectable queries and a chart, not a guessed text answer.
- 1,626 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Data analysis runs through a configured language model, which receives the content of files and database queries.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/antvis-avaDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Data analysis runs through a configured language model, which receives the content of files and database queries.
This is third-party code. Review the repository files before installing.
What it does
AVA (AI-native Visual Analytics) is a JavaScript framework for natural-language data analysis that the author compares to PandasAI. The `@antv/ava` package connects files and databases through one interface, builds an answer to a question, and lets you inspect the query itself and the intermediate result, not only the model's final output. Chart-type selection follows the question and the data's shape, rather than being set by hand in advance. For agents, a separate skill installs with `npx skills add antvis/AVA`: it gives the agent this same analysis capability without rewriting the prompt for every project. The library can also be used directly from the CLI without the skill, when only programmatic access is needed.
Who it is for. Developers and analysts who want an inspectable answer to a data question with a chart, not just text from a model.
Good fit when
- A question about a CSV or database needs to become an inspectable query, not a guessed text answer.
- Chart type needs to follow the question and data shape, not be set by hand in advance.
- Files and databases need connecting through one interface without separate code per source.
- An agent needs data analysis as a ready skill, without rewriting a prompt for every project.
Not a fit when
- Python-stack analytics is needed: AVA is written in JavaScript and aimed at JS apps and agents.
- Manual control over every chart matters, without automatic chart-type selection.
- No model access is configured for the analysis: without it, neither the skill nor the library give a meaningful answer.
Example request
Load this sales CSV and show which region generates the most revenue by month, with a chart.Limitations
AVA is written in JavaScript and aimed at JS applications and agents, not a Python data-analysis stack. Answering a question needs configured model access: without it, neither the skill nor a direct library call build a meaningful query. Inspectability means the query and intermediate result can be viewed, but it does not replace a human checking the numbers before they go into a report. Installing the skill with `npx skills add antvis/AVA` needs an agent that supports this skill-install format.
How to disable. Remove the skill from the agent's skills directory, or uninstall the `@antv/ava` package from the project's dependencies (`npm uninstall @antv/ava`).
FAQ
How do I install AVA as an agent skill?
Run `npx skills add antvis/AVA` and follow the install instructions from the agent-skill guide.
Can I see how an answer was built?
Yes, the query itself and the intermediate result can be inspected, not just the final text or chart output.
Does AVA fit Python projects?
No, the library is written in JavaScript and aimed at JS applications and agents.
Do I need to set the chart type by hand?
No, the visualization type is picked automatically based on the question and the data's shape.
Can the library be used without the skill?
Yes, install it with `npm install @antv/ava@4` and call it directly from code if the skill does not fit your agent.
Discussion
No questions or reviews yet.
Related
SkillsAntV Chart Visualization Skills
chart-visualization-skills
Data visualization skills from AntV: choosing a chart type and building it. The repository accepts only AI code. The antv CLI
A team of data analysis agents that takes a dataset from raw rows to a trained model and checks its own conclusions.
A template for data scientists: AGENTS.md, CLAUDE.md and skills for Python, SQL, Polars and model diagnostics.
Official Weaviate skills: vector database search, data import and ready-made RAG app blueprints for Claude Code and Cursor
BI as code: YAML dashboards, a semantic layer and a local MCP server through which an agent drafts charts.
The Zvec skill guides an agent through the embedded vector database: collection schema, hybrid search, RAG and multimodal queries
README in short
The author cites the Sive app, built on AVA, as a working example of what a finished data-analysis app looks like on this framework.