AnalyticsCLI.com
Feed your coding agent with
Analytics Data
Let your coding agent analyze your apps user behavior so it can create data-driven decisions.
Beta is currently free for all users.
Example Use Cases.
Help your AI Agent to make better decisions based on your analytics data.
Let AI access your analytics data via CLI.
Dashboards were yesterday. Automate your product improvements now.
Install the CLI
By installing our CLI and our AI Skill, you give your AI the ability to access your analytics data.
# One-liner installcurl -fsSL https://analyticscli.com/install.sh | bash
// install and initialize
import { initAnalytics } from '/sdk';
const analytics = initAnalytics({
apiKey: 'pk_live_xxx',
app: 'my-mobile-app',
});
// track onboarding progress
analytics.track('onboarding_step_completed');
Integrate our analytics SDK in your codebase
You can use our AI skill for an optimized implementation.
Learn more here.
Your coding agent accesses your analytics data and makes decisions based of it
Use your favourite coding agent for this (like codex, claude code, openclaw).
hey codex, find dropoff points in onboarding
Fetching results from analytics...
Query executed in 14ms
"Based on the 54% dropoff at STEP 2, I'll refactor the registration flow to split it into two steps and simplify the required fields."
Fast setup.
Let your coding agent handle the implementation in your apps codebase.
-
Type safe SDK
-
Instant setup using our AI skill
const analytics = new AnalyticsCLI('YOUR_API_KEY');
// Identify users to track their journey
analytics.identify('user_123', {
plan: 'pro',
channel: 'tiktok'
});
// Track custom events with zero latency
analytics.track('onboarding_completed');
1 min
integration
Frequently Asked Questions
Everything you need to know about AnalyticsCLI.
Common Questions
AnalyticsCLI currently connects seamlessly to PostgreSQL and ClickHouse. We are adding support for more databases in the future.
Yes. We only require read-only access to your database. We never store your raw data, we only send schema metadata and the generated queries to the LLM.
You can choose between OpenAI Codex, Anthropic Claude 3.5 Sonnet, and OpenClaw depending on your preferences for speed, cost, and accuracy.
Not at all. You can ask questions in plain English, like "Show me the churn rate of users who completed onboarding last month", and we will generate the SQL and the chart.
Yes, any chart generated can be pinned to your dashboard, exported as CSV, or shared with your team via a unique link.
You can start with a 3-day free trial to see how it works with your data.