@bicharts/chart-mcp 0.6.95 → 0.6.96
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +11 -1
- package/dist/check-chart.mjs +73 -73
- package/dist/index.mjs +67 -67
- package/package.json +4 -4
- package/skills/bic-charts/SKILL.md +13 -2
package/README.md
CHANGED
|
@@ -1,6 +1,16 @@
|
|
|
1
1
|
# @bicharts/chart-mcp
|
|
2
2
|
|
|
3
|
-
> **Make Generative AI for Charts *Better*.**
|
|
3
|
+
> **Make Generative AI for Charts and Maps *Better*.**
|
|
4
|
+
|
|
5
|
+
**AI Data Components for Power BI, Excel, Fabric Apps and the web:** a bivariate map with its two-way key, a what-if
|
|
6
|
+
projection with its controls inside the chart, a flow map, and about 90 everyday chart types. Each one arrives with
|
|
7
|
+
selection, cross-filtering and the legend already wired, so your agent doesn't build them. The engine fits it to your
|
|
8
|
+
whole result at its grain, not a sample, and the code it hands back is yours.
|
|
9
|
+
|
|
10
|
+
Generating code is becoming cheap; generating dependable behavior is still expensive. In a
|
|
11
|
+
[published side-by-side build of a Microsoft Fabric App](https://bizintelligencechampions.com/articles/fabric-app-two-mcp-servers?loc=npmmcp)
|
|
12
|
+
(same model, CLI and template), the agent with RegiaBI got 3 of 3 requested charts right on the first build against
|
|
13
|
+
0 of 3, with 37% less agent time, 72% fewer output tokens and 71% fewer tool calls.
|
|
4
14
|
|
|
5
15
|
**Docs and examples:** [RegiaBI for developers](https://bizintelligencechampions.com/regiabi/developers?loc=npmmcp)
|
|
6
16
|
· live demos in [React](https://codexguy.github.io/bicharts-react-demo/) and
|