Agent skills for the VisActor visualization ecosystem. Create data stories, editorial image series and biography posters with independently installable Skills. VChart and VTable Skills are maintained in their respective library repositories.
| Skill | Tasks |
|---|---|
| data-storytelling | Data analysis, narrative design, scrollytelling, HTML reports and Web-PPT |
| editorial-magazine | Editorial HTML / PNG series for news, biographies, outfits, travel, chronological stories and explainers |
| life-chronicle | Biography posters with photo walls, center grids, photo rings and editable timelines |
Chart and table Skills are maintained in their library repositories. Install them directly from the source:
| Skill | Tasks | Repository |
|---|---|---|
| VChart Skill | Chart generation, diagnosis, image replication and React integration | VisActor/VChart |
| VTable Skill | List tables, pivot tables/charts, interactions and diagnosis | VisActor/VTable |
Each skill includes its instructions, local references, templates and scripts. No separate VisActor CLI, MCP server, model key or hosted VisActor API is required. The host agent supplies the model.
Recommended: use the Skills directly in Codex or Claude Code. Your agent provides model access and tool execution. Other compatible agents may also use these Skills, but do not have dedicated setup instructions here.
Run these commands in the project where you want to use the Skills:
npx skills add VisActor/skills --skill data-storytelling
npx skills add VisActor/skills --skill editorial-magazine
npx skills add VisActor/skills --skill life-chronicleFor chart and table tasks, install the related Skills from their owning repositories:
npx skills add VisActor/VChart --skill vchart-development-assistant
npx skills add VisActor/VTable --skill vtable-development-assistantFollow the installer's prompts to select Codex or Claude Code. You can install each Skill independently. More installation options are documented in the Skills CLI reference.
Requirements:
- Node.js and npm for installation; a host agent with filesystem/shell tools and its own model access.
- A browser and network access to preview generated HTML and load external assets.
| Skill | Runtime requirements |
|---|---|
| data-storytelling | Python 3.10+ (standard library) and Node.js for JavaScript checks; pinned VSeed, VChart, VTable and GSAP load from unpkg |
| editorial-magazine | Python 3.10+, dependencies in its requirements.txt, and Chromium/Chrome for PNG export; font CDN with system fallbacks |
| life-chronicle | Python 3.10+, dependencies in its requirements.txt, and Chromium/Chrome for PNG export; pinned VInfo and font CDN, see environment setup |
Image search, generation and cutout capabilities come from the host; follow the selected Skill's asset requirements. Magazine templates contain placeholders to replace before delivery. See the owning repositories for VChart/VTable Skill requirements.
After installation, ask your agent to use the selected Skill with your data. For example:
- “Use data-storytelling to explain retention changes in this CSV to our growth lead as a shareable HTML story.”
- “Use editorial-magazine to turn this article and its sources into an editable HTML and PNG image series.”
- “Use life-chronicle to create a biography poster from these sourced events and photos, with editable timeline text.”
Data-storytelling uses the current agent to author VSeed configurations and local scripts to assemble the output; it needs no hosted chart generation service. Images use host capabilities or the declared fallback. Web-PPT is a browser presentation, not a .pptx file.
Generated JavaScript/HTML is executable code. Preview your own or reviewed artifacts locally; the test browser is not a sandbox for arbitrary untrusted submissions.
The example gallery includes a CO₂ scrolling story and a Las Vegas Web-PPT, with prompts, available inputs, final artifacts and previews.
This section is for contributors editing Skills or running local tests. The default and primary maintenance branch is develop. Start with CONTRIBUTING.md.
| Path | Contents |
|---|---|
skills/<skill-name>/ |
Independently installable runtime: SKILL.md, references, templates/assets and scripts |
tests/<skill_name>/ |
Skill-specific deterministic Python regressions; hyphens become underscores for Python imports |
tests/test_* |
Shared tooling/installation checks and Node.js browser/Story integration tests |
examples/<skill-name>/<case-name>/ |
Curated examples with prompts, inputs, results and previews |
cases/<skill-name>/<case-name>/ |
Fixed Agent prompts, inputs and acceptance guidance |
tests/browser/cases/<skill-name>/<case-id>/ |
Fixed browser cases: configuration, inputs and expectations |
tests/browser/scripts/ |
Browser smoke builds and checks; dependencies are pinned in tests/browser/ |
.cache/skill-dev/<skill-name>/<run-name>/ |
Ignored local workspaces, outputs, screenshots and logs |
docs/ |
Contributor references; docs/migration/ contains historical migration records |
Keep each runtime Skill self-contained. It must not depend on tests/, cases/, examples/ or another Skill's local directory. See contribution conventions for adding Skills and selecting checks.
From the root of your local repository, install the working copy of a Skill:
npx skills add ./skills/data-storytelling
npx skills add ./skills/editorial-magazine
npx skills add ./skills/life-chronicleAlternatively, copy the complete Skill directory into your agent's skill directory, including its references, templates and scripts. Do not copy only SKILL.md. See the local debugging guide for Codex and Claude Code workspace setup. Refresh the installed copy after changing a Skill before retesting.
The Python maintenance environment requires Python 3.12+ and uv:
uv sync --locked
uv run pytest tests -q
uv run ruff check skills testsFixed browser cases live under tests/browser/cases/, with configuration, inputs and acceptance criteria together. After the browser setup in that guide, run uv run python tests/browser/scripts/smoke.py, or select --skill data-storytelling / --case story-ppt. These checks do not call a model.
Use Codex, Claude Code or another compatible mature agent to run a fixed task or your own prompt and inputs. Follow local debugging to prepare an ignored .cache/skill-dev/<skill-name>/<run-name>/ workspace and retain artifacts, screenshots and available logs. This repository does not provide a model API harness or unified agent runner.
Fixed tasks currently cover data-storytelling; browser smoke covers its HTML stories and Web-PPT. Editorial-magazine and life-chronicle have script regressions, but no fixed tasks or browser smoke yet. Review generated results for data accuracy, task requirements and interactions; different correct visual designs are acceptable. Agent tasks are not run in CI.
Skills maintained here are versioned together at repository level. Each formal Skills release is identified by a Git tag covering the Skill instructions, references, templates and scripts. VChart and VTable Skill versions are managed by their respective repositories.
Ordinary edits are identified by Git commits; each Skill change does not require a release tag.
Use a release tag or commit SHA to identify the version when reporting an issue or reproducing a result; branch contents can change. Dependency versions such as VTable 1.26.8 identify the underlying library, not the Skill release. Historical VMind SDK tags remain unchanged and belong to the earlier SDK.
The previous VMind SDK is no longer maintained. Its code and documentation remain in historical releases.