Send to Salesforce agent
Same model, with and without SignalPilot.
SignalPilot96.9%
Snowflake CoCo
65.1%
dbt Wizard
58.1%
Claude Code alone
39.5%
Both runs: Claude Code with Sonnet 4.6. #1 on the leaderboard · every transcript public →
#1 on ADE-Bench by dbt LabsBuild a Data Context Factory
For Agentic Analytics, Decisions, Actions
SignalPilot curates business data context and evals as code, built from your data models, BI tools, and agent traces. It improves with every interaction, so your teams and our multiplayer agents can take confident actions.
SignalPilot · Context Factory
Agent chatMarcus · RevOps
What was churn last quarter?
churn· not in your context yet
No churn definition yet, so I drafted one for your data team.
Add a churn definitionIn review
0
Data team reviewContext as code
Define churnIn review
semantic-models/customer.yml+4
Approve and merge
Approved and merged by Hannah, Data
Built by people who live in the data stack
You centralized your data.
But your context is scattered and uncontrolled.
Your agent gives unpredictable, often wrong answers querying your warehouse because it doesn't understand your business's context and ontology.
Context is scattered. Across dbt, BI tools, agent traces, and people's heads.
Context layers go stale. Faster than you can edit them manually.
Data teams become the bottleneck. Every quick question waits in their queue.
Teams don't trust the numbers. Slowing agent adoption.
What was churn last quarter?asked 5 times
"Before SignalPilot, I told my team that if the number is not from one of the trusted dashboards we maintain, don't assume the number from your agent is right."
The Context FactoryOne source of truth
for multiplayer agentic work
Everyone sells you a context layer, but nobody maintains it
Manually writing definitions, semantic layer, and skill files doesn't scale. We draft your context, metrics, and evals as code from dbt, BI tools, and agent traces, and keep them accurate as your business changes.
Every data context checked before it ships
Every push to GitHub gets its blast radius mapped and its evals run, so you sleep better at night while your team and their agents ship new context.
feat: refund reason on int_orders_with_refunds #212
signalpilotproposed · feat/refund-reason → main
- add refund_reason to int_orders_with_refunds
- fix: keep 1:1 grain on order_id · signalpilot
- ✓Dana R. approved these changes
signalpilot / evalsqueued
Waiting on checks.Merging is blocked
Claude Code
60.5% came back wrong
39.5%
Claude Code + SignalPilot
3.1% came back wrong
96.9%
The sandboxed agentA better, multiplayer agent
at a fraction of the cost
Sandboxed cloud agent for every chat, every number backed by evidence
Everyone on your team gets answers with the evidence attached, and each question runs on the cheapest model that gets it right.
Access from anywhere, without a zillion-step setup
Keep the harness you like. Reach SignalPilot from Claude Cowork, Codex, and more through our MCP app. Your data team sets it up once; every task runs in a cloud sandbox.
"You guys nailed it! This helps the execs go from a static dashboard to answering the 'so what' question in minutes, without adding to the backlog of our analytics engineers."
Claude Opus 5.5
$1.20 per run
$1,200
GLM 5.3 on SignalPilot
$0.12 per run
$120
#1 in BenchmarksSelf-improving context, powering the #1 data agent
SignalPilotThe hardest public data-engineering benchmark.
SignalPilot*75%
Databao · JetBrains
60.29%
Shadowfax
41.18%
Spider Ext · GPT-5
39.71%
*75% submitted in August; official leaderboard shows 65.6% (May).
Who it is for
Built for the people who own the number.
FAQ
Questions we hear on every call.
Short answers. The long ones are a demo away.
Is this another agent?
No. The sandboxed cloud agent is only a part of the whole offering. SignalPilot is primarily a development lifecycle management platform for your data context and agents: a Context Factory that keeps your context, metrics, and evals correct, plus a cloud sandbox for the agents that use them. Once set up, you can use these sandboxed agents from Claude Cowork, Codex, or Slack.
Why not just use Claude Code?
Claude Code makes one person faster. Data work at a company is mostly alignment of controlled business vocabulary and procedures. SignalPilot is multiplayer. Teams settle definitions in chat, your data team merges them, and every person and agent answers from the same agreed context.
Do we have to document everything first?
No. Manually writing definitions, semantic layer, and skill files doesn't scale, so we ingest your dbt models, BI tools, dashboards, and agent traces and draft the context for you. Your team curates the draft, and we help you keep it accurate as your business changes.
Which warehouses?
Snowflake, Databricks, BigQuery, Postgres, DuckDB, Redshift, and more.
Where does our data go?
Queries run in a SignalPilot cloud sandbox against your warehouse, on your warehouse's own permissions. Definitions live in dbt, Apache Ossie, and Markdown in your repo. Self-host with Docker Compose if you prefer.
What does it cost?
From $100 a month for a five-seat team, then $0.50 per successful thread and $0.01 per governed query. See pricing →
See it on your stack.
We will trace one workflow your leadership already reviews and show you the result with its evidence, on your data.
Open source · Apache 2.0 · GitHub · Slack