chart-visualization-skills
AGENTS.md502 starsChanged 52 days agoMITView on GitHub
This file defines the maintenance principles and constraints for agents modifying this repository. It is not a project introduction or a setup guide. Whenever a skill is added, modified, or removed under skills/, keep the following in sync in the same change:
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository10 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- antv-g2-chartSkill
- antv-g6-graphSkill
- antv-gpt-visSkill
- antv-infographicSkill
- antv-t8-ntvSkill
- antv-x6-editorSkill
- chart-visualizationSkill
- gpt-visSkill
- icon-retrievalSkill
The file
We split this AGENTS.md into the rules it gives an agent
11 rules. About 0 are common, 5 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Keep it simple. Prefer editing existing files.
- Distinctive: line 9
Keep code and docs in sync.
- Very specific: line 10
Write for retrieval quality.
- Very specific: line 14
After modifying skill documents under src/content/, rebuild the index (npm run build:index) and run tests (npm test).
- Very specific: line 15
Do not introduce new runtime dependencies unless strictly necessary.
- Distinctive: line 16
Do not change the zvec index field structure or embedding dimension (512d) unless the build script, retriever, and all tests are updated …
- Distinctive: line 17
Write commit messages in English.
- Very specific: line 25
Every skill directory under skills/ must be listed under ## Available Skills.
- Distinctive: line 27
Add entries for new skills, update entries for changed skills, and remove entries for deleted skills.
- Very specific: line 31
Every skill directory under skills/ must have a corresponding entry in the plugins array, in this format:
- Very specific: line 43
Add entries for new skills, update description for changed skills, and remove entries for deleted skills.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AGENTS.md
This file defines the maintenance principles and constraints for agents modifying this repository. It is not a project introduction or a setup guide.
Principles
skills/is the single source of truth for chart generation knowledge. All changes to chart usage, best practices, and constraints belong in skill documents, not scattered across code or other docs.- Keep it simple. Prefer editing existing files; do not add new files or abstraction layers unless necessary.
- Keep code and docs in sync. When changing skill content or retrieval logic, update the affected indexes, tests, and docs in the same change.
- Write for retrieval quality. Skill frontmatter fields (
title,description,tags,use_cases,anti_patterns) are retrieval fields — write them so the skill can be found accurately, not just read by humans.
Constraints
- After modifying skill documents under
src/content/, rebuild the index (npm run build:index) and run tests (npm test). - Do not introduce new runtime dependencies unless strictly necessary; prefer existing dependencies (
@antv/context,commander,gray-matter). - Do not change the zvec index field structure or embedding dimension (512d) unless the build script, retriever, and all tests are updated accordingly.
- Write commit messages in English.
Keeping Docs and Config in Sync
Whenever a skill is added, modified, or removed under skills/, keep the following in sync in the same change:
1. README.md — "Available Skills" section
- Every skill directory under
skills/must be listed under## Available Skills. - Each entry must include an emoji icon, the skill name in bold (matching the directory name), a one-line description matching the skill's
SKILL.mdfrontmatterdescription, and a short paragraph elaborating on its capabilities. - Add entries for new skills, update entries for changed skills, and remove entries for deleted skills. Preserve the section's existing formatting style.
2. .claude-plugin/marketplace.json — plugins array
- Every skill directory under
skills/must have a corresponding entry in thepluginsarray, in this format:
{
"name": "skill-name",
"description": "Description from SKILL.md frontmatter.",
"source": "./",
"strict": false,
"skills": ["./skills/skill-name"]
}- Add entries for new skills, update
descriptionfor changed skills, and remove entries for deleted skills. Keep the JSON valid.
Checklist after any skill-related change
- All skill directories in
skills/are listed inREADME.mdunder## Available Skills - All skill descriptions in
README.mdmatch the correspondingSKILL.mdfrontmatter - All skill directories in
skills/have a corresponding entry in.claude-plugin/marketplace.json - All
descriptionfields inmarketplace.jsonmatch the correspondingSKILL.mdfrontmatter - No stale entries exist in either
README.mdormarketplace.jsonfor removed skills marketplace.jsonremains valid JSON after changes
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

