Imagine handing a new developer an SDK, a stack of docs, and a deadline. They read for twenty minutes, write ten lines, break something, go back to the docs, second-guess a prop name, and try again.
Now multiply that by every component they'll embed, every agent, every color scheme, and every project they'll touch.
That gap, between "I know what I want this to look like" and "I know the exact syntax to make it happen," is the tax every developer pays on embedded analytics.
SpotterCode has been closing that gap in AI-native IDEs like Cursor, Claude Code, and more for a while now, connecting a developer’s editor to ThoughtSpot's SDK and API knowledge through an MCP server. Now, it’s also available in the Developer Playground, right where developers already go to prototype before they build.
What It Does
In the Developer Playground, pick a component and type what you want:
“Embed a Liveboard”
“Embed Spotter and rename the agent to Ava”
“Make the header teal to match our brand”
SpotterCode generates best-practice code against ThoughtSpot’s actual SDK and REST APIs, then drops it in the editor. Hit ‘Run’ to see it in the preview panel, and if it detects a conflict with existing code, it flags it instead of overwriting your work. Keep chatting to refine further, or edit the code directly whenever you'd rather take over.
A build-test-fix cycle across docs, editors, and browser tabs becomes a single conversation: you describe the outcome, watch it render, and only touch code when you actually want to.
Why Grounded Agents Win
Syntax is rarely the hardest part of embedding a component. What slows developers down is knowing which prop controls what, which endpoint does what, and what the docs assume you already know. Closing that gap takes more than a coding agent. It takes one grounded in the platform's documented surface.
That grounding is what's missing as AI adoption outpaces trust. Usage is up, but confidence in AI output has slipped. The top complaint isn't wrong code. It's code that's almost right, the plausible mistakes an agent makes when it's guessing. The fix isn't a bigger, more autonomous agent. It's a narrower one, scoped to known territory instead of improvising around it.
That's the shift SpotterCode is built for. It uses MCP to work with ThoughtSpot's actual SDK and APIs instead of guessing.
IDE or Developer Playground?
The IDE is where you build. SpotterCode reads your project structure, generates code that fits your app's architecture and design system, and keeps your implementation aligned as SDKs and APIs evolve.
The Developer Playground is where you explore. A lot of embedding work isn't writing code from scratch. It's trying a component, seeing how it renders, adjusting a label or a color, trying again. That loop runs faster somewhere disposable, with a live preview sitting two inches from the prompt. Once you've settled on what you want, the code comes with you.
Getting Started
SpotterCode is available in the Developer Playground now. Open the Playground in your ThoughtSpot instance to try it, or see the SpotterCode documentation for more details.
New to ThoughtSpot? Start a free trial and see how fast you can build, white-label, and deploy embedded analytics with your own data.
