Permettez aux agents IA de créer des graphiques expressifs, interactifs et de haute qualité à partir de spécifications simples et modifiables par l'humain via plusieurs moteurs de rendu.
Aussi connu sous microsoft/flint-chart, flint-chart-mcp
- Étoiles
- 4 579
- Outils
- 9
- Téléchargements, 30 jours
- —
- Dernière version
- —
- Dernier commit
- 15 nov. 2026
- Licence
- MIT
Qu'est-ce que Flint MCP ?
Permettez aux agents IA de créer des graphiques expressifs, interactifs et de haute qualité à partir de spécifications simples et modifiables par l'humain via plusieurs moteurs de rendu. Il expose 9 outils, comme render_chart, compile_chart, validate_chart.
Résumé généré à partir du README du projet
À propos du MCP server Flint
Flint est un langage de visualisation conçu pour aider les agents IA à générer des graphiques interactifs de haute qualité à l'aide de spécifications simples. Il fonctionne comme un MCP server qui s'intègre à des clients tels que Claude, Cursor et GitHub Copilot dans VS Code. Les utilisateurs peuvent l'installer localement via npx ou se connecter à un serveur HTTP hébergé. Le système permet aux agents de produire des visualisations interactives, des fichiers statiques PNG ou SVG, ou des spécifications natives au backend directement au sein d'une conversation.
La plateforme prend en charge 52 types de graphiques différents répartis sur cinq moteurs de rendu distincts, notamment Vega-Lite, ECharts, Chart.js, Plotly et les graphiques natifs Excel. Elle simplifie le processus de création en utilisant des types sémantiques pour gérer automatiquement les échelles, les axes et le formatage. Les développeurs peuvent également utiliser la bibliothèque npm flint-chart pour intégrer ces graphiques dans des applications React ou JavaScript pur, en utilisant 14 thèmes prédéfinis et 23 préréglages d'interaction pour maintenir une cohérence visuelle.
Ce que vous pouvez faire avec
- 1Générer des graphiques financiers interactifsDemandez à un agent de récupérer des données boursières et de les afficher en utilisant un style professionnel spécifique avec des effets au survol.
- 2Construire des tableaux de bord de données avec ReactIntégrez la bibliothèque Flint dans une application web pour restituer des graphiques complexes à partir d'objets de données.
- 3Créer des rapports Excel modifiablesUtilisez l'agent pour produire des modèles de graphiques pouvant être utilisés directement dans des logiciels de tableur.
- 4Appliquer des thèmes de marque cohérentsAppliquez une identité visuelle unifiée à travers plusieurs types de graphiques différents en utilisant une seule spécification de thème.
Extrait du README du projet. Chaque cas d'utilisation est vérifié par rapport à une citation du document.
Comment installer Flint MCP
Formats vérifiés le 10 nov. 2026Ajouter à CursorAjouter à GooseInstaller Flint MCP dans Claude Code
$ claude mcp add flint-chart -- npx -y flint-chart-mcpInstaller Flint MCP dans Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"flint-chart": {
"command": "npx",
"args": [
"-y",
"flint-chart-mcp"
]
}
}
}Installer Flint MCP dans VS Code
.vscode/mcp.json
{
"servers": {
"flint-chart": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"flint-chart-mcp"
]
}
}
}Installer Flint MCP dans VS Code CLI
$ code --add-mcp '{"name":"flint-chart","type":"stdio","command":"npx","args":["-y","flint-chart-mcp"]}'Installer Flint MCP dans Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"flint-chart": {
"command": "npx",
"args": [
"-y",
"flint-chart-mcp"
]
}
}
}Installer Flint MCP dans Codex
$ codex mcp add flint-chart -- npx -y flint-chart-mcpInstaller Flint MCP dans Codex config.toml
~/.codex/config.toml
[mcp_servers.flint-chart] command = "npx" args = ["-y", "flint-chart-mcp"]
Installer Flint MCP dans Gemini CLI
$ gemini mcp add -s user flint-chart npx -- -y flint-chart-mcpInstaller Flint MCP dans Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"flint-chart": {
"command": "npx",
"args": [
"-y",
"flint-chart-mcp"
]
}
}
}Démarré par HostedMCP le 16 nov. 2026
Nous avons lancé cette commande pour flint-chart-mcp 1.0.1 dans un conteneur isolé, sans réseau ni clé. Il a répondu et listé 9 outils.
Outils MCP Flint 9
Lu sur le serveur avec tools/listrender_chartCompile a Flint chart spec for a backend and render it to a STATIC image (PNG) or vector (SVG), returned inline. Rendering is local/in-process. The chartjs backend supports PNG only. Prefer validate_chart and show instead when the host supports MCP App UIs; use render_chart for a static artifact or when no App UI is available.
compile_chartCompile a Flint chart spec to a backend-native spec object (Vega-Lite / ECharts / Chart.js) without rendering. Returns the spec JSON, assembly warnings, and the computed layout size.
validate_chartValidate a Flint chart spec for a backend without rendering. Reports whether it is valid, all warnings/errors, and the computed layout size. With an interaction_spec it also reports the entries the chart would drop. With updates it rejects a malformed layer and reports, op by op, whether each will apply (✓) or why not (✗), in the same report show returns. Call it before show and fix what it reports; nothing is kept between calls.
list_chart_typesList the available chart types, their encoding channels, and the interaction presets each supports in interaction_spec (Vega-Lite only), for a backend, or for all backends when none is given.
list_interaction_presetsList the interaction presets for interaction_spec: what the reader does, what each needs from the chart, its reset gestures, and every option it accepts under `options`. Pass `chartType` for the presets a Vega-Lite chart type supports, or `type` for one preset. Which presets answer a request and how they combine is in get_flint_skill with skill "interaction", section "Add interactions to the chart".
get_flint_skillLecture seuleReturn a bundled Flint skill as markdown. Read it before you write the matching spec: `chart` for a ChartAssemblyInput, `theme` for a ThemeSpec, `interaction` for an interaction_spec, ChartUpdate layers, reading a mounted chart, linking charts, or a bespoke interaction; `board` for a board spec: headline, chart choice, layout, links. Pass `section` with one "## " heading to get only that part; an unknown section lists the headings. The same text is served as the flint://agent-skill, flint://theme-skill, flint://interaction-skill, and flint://board-skill resources.
list_themesList Flint visual theme presets for `theme_spec` (for example, `theme_spec: "economist"`). Pass an `id` for preset-specific authoring guidance. To customize a preset, use an object with `extends` and a small set of overrides.
validate_boardCheck a board of several Flint charts without drawing it: the report gives the layout, each chart (✓/✗ per update op), what each chart sends and receives under link_spec, and a review of what the board should still fix. Nothing is kept: fix the spec, check it again if needed, then pass the same board to show. Before the first board, call get_flint_skill with skill "board".
showPut a chart or a board on screen: pass the full spec as chart or as board, after validate_chart or validate_board report nothing left to fix. The card draws it live with an editing panel; the result is the same report. A spec with errors draws nothing. Nothing is kept: to change what is on screen, change the spec and call show again; a new card shows it and earlier cards stay as they were.
| render_chart | Compile a Flint chart spec for a backend and render it to a STATIC image (PNG) or vector (SVG), returned inline. Rendering is local/in-process. The chartjs backend supports PNG only. Prefer validate_chart and show instead when the host supports MCP App UIs; use render_chart for a static artifact or when no App UI is available. | |
| compile_chart | Compile a Flint chart spec to a backend-native spec object (Vega-Lite / ECharts / Chart.js) without rendering. Returns the spec JSON, assembly warnings, and the computed layout size. | |
| validate_chart | Validate a Flint chart spec for a backend without rendering. Reports whether it is valid, all warnings/errors, and the computed layout size. With an interaction_spec it also reports the entries the chart would drop. With updates it rejects a malformed layer and reports, op by op, whether each will apply (✓) or why not (✗), in the same report show returns. Call it before show and fix what it reports; nothing is kept between calls. | |
| list_chart_types | List the available chart types, their encoding channels, and the interaction presets each supports in interaction_spec (Vega-Lite only), for a backend, or for all backends when none is given. | |
| list_interaction_presets | List the interaction presets for interaction_spec: what the reader does, what each needs from the chart, its reset gestures, and every option it accepts under `options`. Pass `chartType` for the presets a Vega-Lite chart type supports, or `type` for one preset. Which presets answer a request and how they combine is in get_flint_skill with skill "interaction", section "Add interactions to the chart". | |
| get_flint_skill | Return a bundled Flint skill as markdown. Read it before you write the matching spec: `chart` for a ChartAssemblyInput, `theme` for a ThemeSpec, `interaction` for an interaction_spec, ChartUpdate layers, reading a mounted chart, linking charts, or a bespoke interaction; `board` for a board spec: headline, chart choice, layout, links. Pass `section` with one "## " heading to get only that part; an unknown section lists the headings. The same text is served as the flint://agent-skill, flint://theme-skill, flint://interaction-skill, and flint://board-skill resources. | Lecture seule |
| list_themes | List Flint visual theme presets for `theme_spec` (for example, `theme_spec: "economist"`). Pass an `id` for preset-specific authoring guidance. To customize a preset, use an object with `extends` and a small set of overrides. | |
| validate_board | Check a board of several Flint charts without drawing it: the report gives the layout, each chart (✓/✗ per update op), what each chart sends and receives under link_spec, and a review of what the board should still fix. Nothing is kept: fix the spec, check it again if needed, then pass the same board to show. Before the first board, call get_flint_skill with skill "board". | |
| show | Put a chart or a board on screen: pass the full spec as chart or as board, after validate_chart or validate_board report nothing left to fix. The card draws it live with an editing panel; the result is the same report. A spec with errors draws nothing. Nothing is kept: to change what is on screen, change the spec and call show again; a new card shows it and earlier cards stay as they were. |
Vérifications automatisées
Exécuté le 16 nov. 2026Rapport complet →A passé nos vérifications automatisées. Il ne s'agit pas d'un audit de sécurité : lisez le code et donnez au serveur l'accès le plus restreint dont il a besoin. Fonctionnement des vérifications
- Dépôt source lié et accessiblemicrosoft/flint-chart is public and reachablegithub.com/microsoft/flint-chart ↗Réussi
- Le nom du package correspond au dépôt sourceflint-chart-mcp names microsoft/flint-chart as sourcenpmjs.com/package/flint-chart-mcp ↗Réussi
- Provenance du build (attestation Sigstore)Sigstore attestation for flint-chart-mcpnpmjs.com/package/flint-chart-mcp/v/1.0.1#provenance ↗Réussi
- Licence open source (OSI)MITspdx.org/licenses/MIT.html ↗Réussi
- Commit ou version publiée au cours de l'année écouléeLast commit or release 2026-11-15github.com/microsoft/flint-chart ↗Réussi
- OpenSSF ScorecardNon vérifié
- Nom du package non proche d'un package populaireNo popular package one edit away from flint-chart-mcpRéussi
- Aucun avis critique dans le package ou ses dépendancesNon vérifié
- Aucune instruction cachée destinée au modèle9 tool descriptions and the README read; no override phrases, credential paths or hidden charactersRéussi
- Les outils destructifs sont étiquetésNo tool named like delete, drop or execRéussi
- Le point de terminaison distant répond à la poignée de main MCPNon vérifié
README
Voir sur GitHubProject site · 中文主页 · MCP server
Flint is a visualization language that lets AI agents reliably create expressive, good-looking and interactive charts from simple, human-editable chart specs. It is available as an MCP server for everyday use with your agent, and as an npm library for building interactive applications.
Flint supports 52 chart types across 5 rendering backends — Vega-Lite, ECharts, Chart.js, Plotly, and native Excel charts — through one unified interface, with 14 themes and 23 interaction presets. To create bespoke visualizations, design custom themes in the Theme Lab and build advanced interactions with the Flint API.
https://github.com/user-attachments/assets/c1cda0fa-3ea7-41af-b3fd-a127bc83d641
Updates
- October 19, 2026 — Flint 1.0 introduces Flint Interactive: interaction presets, semantic events, and an interaction API that connects charts to your app. (changelog)
- August 12, 2026 — Flint 0.5.1 extends visual themes to the Plotly backend and adds Vega-Lite Calendar Heatmap.
- August 5, 2026 — Flint 0.5.0 introduces formal visual themes with preset themes, custom ThemeSpec authoring, and preset inheritance. (v0.5.0)
- July 24, 2026 — Flint 0.4.0 adds 38 Plotly chart types and 18 native, editable Excel chart templates. (v0.4.0)
See the changelog for every release.
Get started
In your agent app (MCP server)
Chart right in the conversation with any MCP client, such as GitHub Copilot in VS Code, Claude, Claude Code, Codex, or Cursor.
⚡ Quick start: paste this setup request into your agent and let it configure Flint for the current project. For manual setup, see the MCP server guide.
Help me set up Flint as an MCP server named "flint"!
1. Add it to my client config, using whichever option my client supports:
- a local (stdio) server: npx -y flint-chart-mcp
- the hosted server over HTTP: https://flint.data-formulator.ai/mcp
2. Verify the setup by listing the available Flint chart types.
3. Show me a fun first chart: how many moons each planet has, in a Flint theme you like.
Then ask your agent for a chart, for example:
Go to Yahoo Finance, find today's percentage change for the Magnificent Seven
tech stocks, and compare their performance with Flint in Economist style.
Let me hover to read each stock's change.
The agent can open an interactive chart right in the conversation, return a static PNG or SVG, or produce a backend-native spec.
Build with your coding agent
Are you a developer building agent apps? Use
flint-chart, Flint's npm library,
to bring charts into your apps, dashboards, and data stories:
npm install flint-chart
The library compiles Flint specs into backend-native chart specs and provides React and vanilla JavaScript APIs for embedding interactive charts.
A Flint spec describes the data, semantic types, and field mappings, with
optional theme and interaction settings. In React, pass it to <FlintChart>
to embed an interactive chart:
import type { ChartAssemblyInput } from 'flint-chart';
import { FlintChart } from 'flint-chart/react';
const spec: ChartAssemblyInput = {
data: { values: [
{ Month: '2026-01', Users: 1200 },
{ Month: '2026-02', Users: 1800 },
] },
semantic_types: { Month: 'YearMonth', Users: 'Quantity' },
chart_spec: {
chartType: 'Line Chart',
encodings: { x: 'Month', y: 'Users' },
},
theme_spec: 'economist',
interaction_spec: { interactions: [{ type: 'inspect-index' }] },
};
<FlintChart spec={spec} />
You can copy the following instruction to your coding agent to build a chart with Flint, then adapt it to your data and app:
Read Flint's chart authoring and interaction skills:
https://github.com/microsoft/flint-chart/blob/main/agent-skills/flint-chart-author/SKILL.md
https://github.com/microsoft/flint-chart/blob/main/agent-skills/flint-interaction-author/SKILL.md
Use these skills to add a chart to this app.
Install flint-chart and the renderer dependencies needed for this project.
Plot monthly users by region from our data in Economist style.
Let readers hover to compare regions at each month, and add a region selector
that updates the chart.
Integrate it with the app's existing layout and verify it works.
The agent can also generate backend-native specs or editable Excel charts for reports and documents. See the API reference, interaction API, and live editor for the underlying APIs and examples.
How it works
A Flint spec consists of the data, semantic types, the chart spec, and optional theme and interaction specs. Instead of relying on complex user-side logic to glue them together, the compiler resolves each part against the others and composes a complete backend-native spec.
1. Describe data with semantic types
Fields are described with semantic types such as Rank, YearMonth, Delta,
or Temperature, along with their domains and units. The compiler uses them to
infer parsing, scales, axes, formatting, and colors — a short Flint spec becomes
a complete backend spec.
2. Build charts once, render on any backend
Charts are defined with high-level chart types and encodings. The compiler adapts sizing, spacing, labels, and legends to the data, for example as it gets denser, and the same spec renders across Vega-Lite, ECharts, Chart.js, and Plotly, or as native Excel charts. See the API reference and the gallery.
3. Unify the look with a theme spec
Geometry, typography, and color are reusable formal specs that apply across a whole chart library, independent of the data and chart type. Pick one of 14 preset themes, such as the New York Times, the Economist, Swiss, or McKinsey, or inherit one and override only what belongs to your brand:
theme_spec: { extends: 'economist', id: 'our-brand', ink: { series: { single: '#6b3fa0' } } }
Compare every preset in the theme explorer, design your own in the Theme Lab, and see the theme guide for the full reference.
4. Add interactions with an interaction spec
Flint provides 23 interaction presets and a programming API for building interactive charts into applications. The compiler resolves user interaction events into semantic information, so your app only answers with high-level chart updates:
import { FlintChart } from 'flint-chart/react';
<FlintChart
spec={{ ...input, interaction_spec: { interactions: [{ type: 'brush-x' }] } }}
onChange={({ phase, state }) => {
if (phase === 'commit') table.filter(state.selected);
}}
/>
Try the presets in the interaction gallery, see bespoke ones in advanced interactions, and read the interaction spec and programming API.
Documentation
- Project site — gallery, live editor, themes, and interactions
- API reference and backend references for Vega-Lite, ECharts, Chart.js, Plotly, and Excel
- Theme guide, interaction spec, and interaction programming API
- MCP server guide and agent skills
- Paper on arXiv
A Python port is in preview as source in packages/flint-py; its package is not yet released.
Contributing
Contributions are welcome — especially new chart templates and rendering backends. See CONTRIBUTING and the development guide.
git clone https://github.com/microsoft/flint-chart
cd flint-chart
npm install # workspaces: packages/flint-js, packages/flint-mcp, site
npm run typecheck # typecheck flint-chart and flint-chart-mcp
npm run test # Vitest for both packages
npm run build # build both packages
npm run site # project site, gallery, and editor at http://localhost:5274/
Node 18+ is required. The site aliases flint-chart to packages/flint-js/src,
so library edits hot-reload without rebuilding.
flint-chart/
├── packages/
│ ├── flint-js/ npm package `flint-chart` (TypeScript)
│ │ └── src/
│ │ ├── core/ semantics, themes, layout, decisions, shared types
│ │ ├── chart-types/ shared chart definitions and template metadata
│ │ ├── interactive/ interaction presets and runtime
│ │ ├── react/ the `<FlintChart>` React component
│ │ ├── vegalite/ Vega-Lite backend
│ │ ├── echarts/ ECharts backend
│ │ ├── chartjs/ Chart.js backend
│ │ ├── plotly/ Plotly backend
│ │ ├── excel/ native Excel backend
│ │ └── test-data/ fixtures and stress-test generators
│ ├── flint-mcp/ MCP server `flint-chart-mcp`, MCP App UI, and tests
│ └── flint-py/ Python port preview
├── site/ Vite + React project site, gallery, editor, and docs
├── agent-skills/ chart, theme, and interaction authoring skills for agents
├── docs/ user, API, backend, and architecture documentation
└── design-docs/ design proposals and implementation research
This project has adopted the Microsoft Open Source Code of Conduct. See SECURITY.md to report vulnerabilities.
Community
Contributors
Trademarks
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos is subject to those third parties' policies.
License
MIT © Microsoft Corporation
Questions sur Flint MCP
Comment ajouter Flint MCP à Claude Code ?
Exécutez claude mcp add flint-chart -- npx -y flint-chart-mcp dans votre terminal. Ajoutez --scope user pour le rendre disponible dans tous les projets.
Comment ajouter Flint MCP à Cursor ?
Ouvrez ~/.cursor/mcp.json et ajoutez le serveur sous mcpServers, avec la configuration Cursor indiquée dans la section d'installation ci-dessus.
Quels outils Flint MCP fournit-il ?
Il expose 9 outils, dont render_chart, compile_chart, validate_chart, list_chart_types, list_interaction_presets.
Puis-je personnaliser l'apparence des graphiques ?
Oui, vous pouvez choisir parmi 14 thèmes prédéfinis comme ceux de l'Economist ou du New York Times, ou concevoir vos propres thèmes personnalisés via le Theme Lab.
Flint prend-il en charge les graphiques interactifs ?
Oui, il propose 23 préréglages d'interaction et une API de programmation pour intégrer des éléments interactifs tels que le brushing et les sélecteurs dans vos graphiques.