FakerForge

Documentation

Faker Forge Docs

Complete user guides for Data Generation, API Mocking, and MCP integration, organized for both developers and LLM-driven workflows.

Data Generation

Data Assistant

Chat with an agent to shape values, relationships, row counts and exports.

Data Assistant

The Data Assistant is a chat on the schema page. Describe how your data should look in plain language, and it changes the generation settings for you.

Opening the Assistant

Open a parsed schema and click Assistant. The panel slides in from the right with your current conversation.

  • History lists earlier conversations about this schema. Click one to reopen it, or use the trash icon to delete it (this can't be undone).
  • New chat starts a fresh conversation.
  • Press Enter to send and Shift + Enter for a new line.

Conversations belong to one schema and one user. Other users can't see them.

What You Can Ask

  • Shape values: "make 80% of orders completed", "SKUs should look like ABC-1234", "prices between 5 and 500, mostly cheap".
  • Change a whole domain: "customers are based in Kenya and pay in KES", "make it look like a B2B SaaS app".
  • Preview: "show me 5 orders". Sample rows are generated without touching your stored data.
  • Size the run: "100 customers and 1,000 orders".
  • Shape ratios: "about 10 orders per customer". This sets the parent table for orders.
  • Fix relationships: "order_items.order_id should reference orders".
  • Read your SQL: "what CHECK constraints does payments have?". The assistant reads your original uploaded schema.
  • Generate and export: "regenerate the data", "export it as Postgres SQL".

Seeing What It Did

While the assistant works, each step appears as a row (for example Update profile orders.status or Preview rows orders) next to a live loader and timer. After it replies, the steps fold into an "N tool calls" summary; expand a step to see the details or the error it returned.

Relationships

The assistant can add relationships your database doesn't declare, change their type or parent, and remove wrong ones. To keep your data safe:

  • It checks that both columns exist and their types match before adding.
  • Foreign keys declared in your schema can never be removed or re-pointed, only switched between one-to-many and one-to-one.
  • It always describes a change or removal first and waits for your explicit "yes".

See Relationships and Integrity for how relationships are enforced.

What It Can't Change

  • Primary keys, foreign key values, ENUM columns and password columns keep their exact rules.
  • It only works on the schema you have open, and only reads your own schema file.

Applying Changes

Profile, relationship and row-count changes are saved straight away but only reach your stored data when you regenerate. Ask the assistant to regenerate, or use Generate Data.

Usage

Assistant replies use AI tokens from your plan, like other AI features. See Limits and Plans.