ST-1Signal Tracer
An analytics workbench
for humans and agents.
Upload a dataset. Shape it with SQL. Build a chart. The same surface is a tool for an AI agent — so the model works against real data with proper instruments, instead of guessing prose at numbers.
Every transformation is authored SQL. Every chart points at a real column. When an agent does work here, you can see exactly what it did, why, and re-run it yourself.
For humans
A real workbench
Upload CSV or parquet. Profile every column. Cross-filter by clicking a histogram bar. Author derived tables in SQL with autocomplete and live preview. Build charts that honour your data's grain.
For agents
Proper tools, real numbers
An MCP server exposes the same workbench to AI clients over OAuth. Models query, transform, and visualise — they don't paraphrase. Their work lands in your project, with links you can open and inspect.
By design
Visual, traceable, verifiable
SQL is the source of truth. Lineage is explicit. Filters propagate through derived tables and visualisations automatically. Nothing is hidden in a transcript — every step is a row you can read.
Scale
Always snappy. Considered at scale.
Three layers, sized to keep interaction in a single frame and the canonical data honest. The browser does what browsers do well; the server does what servers do well; the seam is explicit.
Up to ~2M rows
Sub-frame interactive
Mosaic precomputes column-summary cubes in DuckDB-WASM. Filter a histogram bar; the table, every other summary, and any chart on top all repaint in the same frame. No round-trips.
2M – 50M rows
Auto-sampled preview
Sources past the threshold materialize as a reservoir sample in your browser for snappy interaction. Every panel reading from the sample carries a clear badge. The canonical data is untouched on the server. Toggle off when you need exact counts.
50M+ rows
Server-side compute
Query the canonical data through the MCP surface, or author a derived rollup that's small enough to explore directly. The browser sees the result; the heavy work stays where the data lives. 1 GiB per upload.
Our position
We think AI ought to work the way a careful colleague works — with real instruments, on real numbers, leaving a trail. Not by reciting plausible-sounding answers. ST-1 is a small bet on that discipline: a quiet workbench where the work is the point.
Get started
Sign in to begin
Authenticate with GitHub. Each project is private to you and shareable by URL. Connect your AI client over MCP from the settings panel.