Question 1

What is Harbinger Explorer?

Accepted Answer

Harbinger Explorer is an AI-powered data workspace that lets you connect any REST API, upload files (CSV, Excel, JSON, Parquet), and explore your data using SQL or natural language — all in your browser. An intelligent agent crawls API documentation, discovers endpoints, and loads data into DuckDB automatically.

Question 2

Do I need technical knowledge?

Accepted Answer

No. The AI agent handles SQL generation, schema mapping, and data loading for you. Just describe what you want in plain English and the agent writes the queries. Power users can also write SQL directly in the Monaco editor with auto-complete.

Question 3

What data formats are supported?

Accepted Answer

Harbinger Explorer supports CSV, TSV, Excel (.xlsx, .xls), JSON, JSON Lines, Parquet, and any REST API with documentation. The AI crawler can read most API doc formats including OpenAPI/Swagger, custom HTML docs, and developer portals.

Question 4

Is my data stored on your servers?

Accepted Answer

No. DuckDB runs entirely in your browser via WebAssembly. Uploaded files are processed client-side in browser memory and never sent to our servers. Only metadata (source names, schemas) is stored in Firestore for your catalog.

Question 5

Can I use it for free?

Accepted Answer

Yes. Every account starts with a 7-day free trial with full access. After the trial, the Starter plan begins at €19/month. No credit card is required to start.

Question 6

Is my data safe?

Accepted Answer

Yes. DuckDB WASM runs entirely in your browser — your data never leaves your machine. PII (personally identifiable information) is auto-detected and pseudonymized before the AI agent sees it. API keys are encrypted with AES-256-GCM and stored server-side only.

Question 7

What happens to uploaded files?

Accepted Answer

Uploaded files are processed in browser memory using DuckDB WASM. The raw file data is never transmitted to or stored on our servers. Only the table schema and metadata are persisted in your catalog.

Question 8

Can the AI see my sensitive data?

Accepted Answer

Columns detected as PII (emails, phone numbers, social security numbers, etc.) are automatically pseudonymized before being sent to the AI agent. The agent works with masked values and never sees the original sensitive data.

Question 9

Where is the infrastructure?

Accepted Answer

The frontend is served via Vercel’s Edge Network for global low-latency delivery. The AI agent runs on Google Cloud Run in the EU (europe-west1). Firebase Auth and Firestore are also hosted in the EU region.

Question 10

Are API keys stored securely?

Accepted Answer

Yes. All API keys are encrypted using AES-256-GCM encryption before storage. Keys are stored server-side only and injected at request time — they never appear in browser storage or client-side code.

Question 11

What APIs can I connect?

Accepted Answer

Any REST API with documentation. The AI crawler reads OpenAPI specs, Swagger docs, custom HTML documentation, and developer portals. Point it at a URL and it extracts endpoints, parameters, and authentication requirements automatically.

Question 12

Can I join data across sources?

Accepted Answer

Yes. DuckDB gives you full SQL power including JOINs, CTEs, window functions, and aggregations across all your loaded tables — whether they came from different APIs, uploaded files, or a mix of both.

Question 13

What chart types are available?

Accepted Answer

Harbinger Explorer supports Bar, Line, Area, and Pie charts powered by Recharts. The AI agent automatically selects the best chart type based on your data and query. You can pin charts to your dashboard for quick access.

Question 14

Can I export my work?

Accepted Answer

Yes. Export your data and queries as CSV, Parquet, SQL scripts, or dbt models. You can also push directly to a GitHub repository. Pro and Team plans unlock additional export formats.

Question 15

Does it work on mobile?

Accepted Answer

Harbinger Explorer has a responsive design that works on mobile devices. However, the full SQL editor and data exploration experience is optimized for desktop workflows with larger screens.

Question 16

Can I cancel anytime?

Accepted Answer

Yes. There are no contracts or lock-in periods. You can cancel your subscription at any time from your account settings. Your data remains accessible until the end of your billing period.

Question 17

What's the difference between Starter and Pro?

Accepted Answer

Starter includes up to 25 sources, 50 crawls per month, agent chat, and CSV/SQL export. Pro unlocks unlimited sources, unlimited crawls, priority crawling, and additional export formats including Parquet and dbt models.

Question 18

Do you offer team discounts?

Accepted Answer

The Team plan at €149/month includes 5 seats with a shared workspace, SSO, and admin controls. Additional seats can be added for €29/month each. Contact us for enterprise pricing with more than 20 seats.

Frequently Asked Questions

Getting Started

Data & Security

Features

Pricing

Still have questions?