QueryWeaver

[new-qw-ui-gif]

Get Started

Docker

💡 Recommended for evaluation purposes (Local Python or Node are not required)

docker run -p 5000:5000 -it falkordb/queryweaver

Launch: http://localhost:5000

Use an .env file (Recommended)

Create a local .env by copying .env.example and passing it to Docker. This is the simplest way to provide all required configuration:

cp .env.example .env# edit .env to set your values, then:docker run -p 5000:5000 --env-file .env falkordb/queryweaver

Alternative: Pass individual environment variables

If you prefer to pass variables on the command line, use -e flags (less convenient for many variables):

docker run -p 5000:5000 -it \  -e APP_ENV=development \  -e FASTAPI_SECRET_KEY=your_super_secret_key_here \  -e GOOGLE_CLIENT_ID=your_google_client_id \  -e GOOGLE_CLIENT_SECRET=your_google_client_secret \  -e GITHUB_CLIENT_ID=your_github_client_id \  -e GITHUB_CLIENT_SECRET=your_github_client_secret \  -e AZURE_API_KEY=your_azure_api_key \  falkordb/queryweaver

APP_ENV=development is what makes the login work on the plain-HTTP http://localhost:5000 this command serves. Drop it (or set anything else) when you put QueryWeaver behind HTTPS, so the session cookie is marked Secure. See Application environment.

Note: QueryWeaver supports multiple AI providers. You can use OPENAI_API_KEY, GEMINI_API_KEY, ANTHROPIC_API_KEY, or AZURE_API_KEY. See the AI/LLM configuration section for details.

For a full list of configuration options, consult .env.example.

Memory TTL (optional)

QueryWeaver stores per-user conversation memory in FalkorDB. By default these graphs persist indefinitely. Set MEMORY_TTL_SECONDS to apply a Redis TTL (in seconds) so idle memory graphs are automatically cleaned up.

# Expire memory graphs after 1 week of inactivityMEMORY_TTL_SECONDS=604800

The TTL is refreshed on every user interaction, so active users keep their memory.

MCP server: host or connect (optional)

QueryWeaver includes optional support for the Model Context Protocol (MCP). You can either have QueryWeaver expose an MCP-compatible HTTP surface (so other services can call QueryWeaver as an MCP server), or configure QueryWeaver to call an external MCP server for model/context services.

What QueryWeaver provides

• The app registers MCP operations focused on Text2SQL flows:

  • list_databases
  • connect_database
  • database_schema
  • query_database

• To disable the built-in MCP endpoints set DISABLE_MCP=true in your .env or environment (default: MCP enabled).

• Configuration

• DISABLE_MCP — disable QueryWeaver's built-in MCP HTTP surface. Set to true to disable. Default: false (MCP enabled).

Examples

Disable the built-in MCP when running with Docker:

docker run -p 5000:5000 -it --env DISABLE_MCP=true falkordb/queryweaver

Calling the built-in MCP endpoints (example)

• The MCP surface is exposed as HTTP endpoints.

Server Configuration

Below is a minimal example mcp.json client configuration that targets a local QueryWeaver instance exposing the MCP HTTP surface at /mcp.

{   "servers": {      "queryweaver": {         "type": "http",         "url": "http://127.0.0.1:5000/mcp",         "headers": {            "Authorization": "Bearer your_token_here"         }      }   },   "inputs": []}

REST API

API Documentation

Swagger UI: https://app.queryweaver.ai/docs

OpenAPI JSON: https://app.queryweaver.ai/openapi.json

Overview

QueryWeaver exposes a small REST API for managing graphs (database schemas) and running Text2SQL queries. All endpoints that modify or access user-scoped data require authentication. In the browser the app uses a signed session cookie established by OAuth or email/password; for CLI and scripts you can use an API token (see tokens routes or the web UI to create one).

Core endpoints

• GET /graphs — list available graphs for the authenticated user
• GET /graphs/{graph_id}/data — return nodes/links (tables, columns, foreign keys) for the graph
• POST /graphs — upload or create a graph (JSON payload or file upload)
• POST /graphs/{graph_id} — run a Text2SQL chat query against the named graph (streaming response)

Authentication

• Add an Authorization header: Authorization: Bearer <API_TOKEN>

Three separate credentials

QueryWeaver keeps its three kinds of "login" independent of one another, so a failure in one never looks like a failure in another:

Because the browser login is a signed cookie, staying logged in costs no database round trip and survives a FalkorDB outage in an already-running process — you keep your session and only the operations that genuinely need the graph fail. Requests that supply an API token explicitly are always checked against the database and are answered with 503 (not 401) when it cannot be reached, so clients retry instead of re-authenticating.

Note the scope: this is about staying logged in, not about booting. QueryWeaver still connects to FalkorDB at startup and will not start without it, so a restart during an outage is not covered.

A browser login lasts 24 hours by default; set BROWSER_SESSION_TTL_HOURS to change that. Logging out clears the session cookie. No API token is issued to the browser, so there is none to revoke — tokens are created explicitly from the tokens API and revoked there. (A legacy api_token cookie left over from an older release is cleared and revoked on logout too.)

Email signup is verified before the account exists

Signing up with an email address and password does not create an account. The submitted details are parked, a six-digit confirmation code is mailed to the address, and the account — and the session — come into being only when that code is typed back into the signup form. So an address the registrant does not control never becomes an account at all, and there is no half-real user for the rest of the system to reason about.

A code rather than an emailed link, because the code has to come back to the session that submitted the form. A link can be opened by anyone who receives it: a stranger could submit your address with a password of their choosing, and your single click would create an account they knew the password to. Nobody can be signed up by someone else here, because the person who fills in the form is the only one who ever holds both halves.

The code is single-use, expires after 15 minutes and tolerates only a handful of wrong guesses before the pending signup is discarded — a short code is only safe while the number of attempts is small. It is also only redeemable in the browser that submitted the form: each submission mints a ticket that stays in that browser's session, and a code presented without its ticket is refused. Entering it signs the browser in directly: the password was chosen minutes earlier, and asking for it again would prove nothing. A code can be re-sent from the same screen, subject to a per-address rate limit; the send budget is per pending signup, so it starts over once the pending signup expires and an address can always be signed up again later. Typing a code that has expired is not one of the wrong guesses and does not discard anything — the pending signup is left where it is so the same screen can send a fresh code.

In development, a message with no mail server configured is written to the application log instead of being sent, so the flow can be completed by copying the code out of the log. This needs APP_ENV=development — anywhere else an unconfigured process refuses the send rather than logging the code and reporting success. Set MAIL_SERVER (plus MAIL_PORT, MAIL_USERNAME, MAIL_PASSWORD, MAIL_DEFAULT_SENDER) to send for real; any provider with an SMTP endpoint works. EMAIL_VERIFICATION_TTL_MINUTES, EMAIL_VERIFICATION_MAX_ATTEMPTS, EMAIL_VERIFICATION_RESEND_SECONDS and EMAIL_VERIFICATION_MAX_SENDS tune the lifetime and the limits. See .env.example for the full list.

The trade-off of a signed session cookie is that it cannot be revoked from the server before it expires: the TTL bounds the damage, and rotating FASTAPI_SECRET_KEY invalidates every browser login at once. API tokens keep their server-side record and so can still be revoked individually and immediately. Shorten BROWSER_SESSION_TTL_HOURS if you need a tighter window.

Examples

1. List graphs (GET)

curl example:

curl -s -H "Authorization: Bearer $TOKEN" \   https://app.queryweaver.ai/graphs

Python example:

import requestsresp = requests.get('https://app.queryweaver.ai/graphs', headers={'Authorization': f'Bearer {TOKEN}'})print(resp.json())

2. Get graph schema (GET)

curl example:

curl -s -H "Authorization: Bearer $TOKEN" \   https://app.queryweaver.ai/graphs/my_database/data

Python example:

resp = requests.get('https://app.queryweaver.ai/graphs/my_database/data', headers={'Authorization': f'Bearer {TOKEN}'})print(resp.json())

3. Load a graph (POST) — JSON payload

curl -H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \   -d '{"database": "my_database", "tables": [...]}' \   https://app.queryweaver.ai/graphs

Or upload a file (multipart/form-data):

curl -H "Authorization: Bearer $TOKEN" -F "[email protected]" \   https://app.queryweaver.ai/graphs

4. Query a graph (POST) — run a chat-based Text2SQL request

The POST /graphs/{graph_id} endpoint accepts a JSON body with at least a chat field (an array of messages). The endpoint streams processing steps and the final SQL back as server-sent-message chunks delimited by a special boundary used by the frontend. For simple scripting you can call it and read the final JSON object from the streamed messages.

Example payload:

{   "chat": ["How many users signed up last month?"],   "result": [],   "instructions": "Prefer PostgreSQL compatible SQL"}

curl example (simple, collects whole response):

curl -s -H "Authorization: Bearer $TOKEN" -H "Content-Type: application/json" \   -d '{"chat": ["Count orders last week"]}' \   https://app.queryweaver.ai/graphs/my_database

Python example (stream-aware):

import requestsimport json
url = 'https://app.queryweaver.ai/graphs/my_database'headers = {'Authorization': f'Bearer {TOKEN}', 'Content-Type': 'application/json'}with requests.post(url, headers=headers, json={"chat": ["Count orders last week"]}, stream=True) as r:      # The server yields JSON objects delimited by a message boundary string      boundary = '|||FALKORDB_MESSAGE_BOUNDARY|||'      buffer = ''      for chunk in r.iter_content(decode_unicode=True, chunk_size=1024):            buffer += chunk            while boundary in buffer:                  part, buffer = buffer.split(boundary, 1)                  if not part.strip():                        continue                  obj = json.loads(part)                  print('STREAM:', obj)

Notes & tips

• Graph IDs are namespaced per-user. When calling the API directly use the plain graph id (the server will namespace by the authenticated user). For uploaded files the database field determines the saved graph id.
• The streaming response includes intermediate reasoning steps, follow-up questions (if the query is ambiguous or off-topic), and the final SQL. The frontend expects the boundary string |||FALKORDB_MESSAGE_BOUNDARY||| between messages.
• For destructive SQL (INSERT/UPDATE/DELETE etc) the service will include a confirmation step in the stream; the frontend handles this flow. If you automate destructive operations, ensure you handle confirmation properly (see the ConfirmRequest model in the code).

Python SDK

The QueryWeaver Python SDK allows you to use Text2SQL functionality directly in your Python applications without running a web server.

Installation

# SDK only (minimal dependencies)pip install queryweaver
# With server dependencies (FastAPI, etc.)pip install queryweaver[server]
# Development (includes testing tools)pip install queryweaver[dev]

Quick Start

import asynciofrom queryweaver import QueryWeaver
async def main():    # Initialize with FalkorDB connection    qw = QueryWeaver(falkordb_url="redis://localhost:6379")
    # Connect a PostgreSQL or MySQL database    conn = await qw.connect_database("postgresql://user:pass@host:5432/mydb")    print(f"Connected: {conn.database_id}")  # "mydb"
    # Convert natural language to SQL and execute — pass the database_id    # returned by connect_database (un-prefixed; namespacing is internal).    result = await qw.query(conn.database_id, "Show me all customers from NYC")    print(result.sql_query)    # SELECT * FROM customers WHERE city = 'NYC'    print(result.results)       # [{"id": 1, "name": "Alice", "city": "NYC"}, ...]    print(result.ai_response)   # "Found 42 customers from NYC..."
    await qw.close()
asyncio.run(main())

Context Manager

async with QueryWeaver(falkordb_url="redis://localhost:6379") as qw:    conn = await qw.connect_database("postgresql://user:pass@host/mydb")    result = await qw.query(conn.database_id, "Count orders by status")# close() runs automatically, awaiting any in-flight background memory writes.

Multiple Instances

Multiple QueryWeaver instances can run side-by-side in the same process. Each holds its own FalkorDB connection and passes it explicitly through every call, so there is no shared global state to collide over.

async with QueryWeaver(falkordb_url="redis://host-a:6379", user_id="tenant_a") as a, \           QueryWeaver(falkordb_url="redis://host-b:6379", user_id="tenant_b") as b:    sales = await a.connect_database("postgresql://user:pass@host-a/sales")    ops = await b.connect_database("postgresql://user:pass@host-b/ops")    await a.query(sales.database_id, "Show top customers")    await b.query(ops.database_id, "Count open tickets")

Available Methods

Advanced Query Options

For multi-turn conversations, custom instructions, or per-request LLM overrides:

from queryweaver import QueryWeaver, QueryRequest
request = QueryRequest(    question="Show their recent orders",    chat_history=["Show all customers from NYC"],    result_history=["Found 42 customers..."],    instructions="Use created_at for date filtering",    # Optional per-request LLM overrides — bypass env-based config    custom_api_key="sk-...",    custom_model="openai/gpt-4.1",)
result = await qw.query("mydb", request)

Handling Destructive Operations

INSERT, UPDATE, DELETE operations require confirmation:

result = await qw.query("mydb", "Delete inactive users")
if result.requires_confirmation:    print(f"Destructive SQL: {result.sql_query}")    # Execute after user confirms    confirmed = await qw.execute_confirmed("mydb", result.sql_query)

Requirements

• Python 3.12+
• FalkorDB instance (local or remote)
• OpenAI or Azure OpenAI API key (for LLM)
• Target SQL database (PostgreSQL or MySQL)

Development

Follow these steps to run and develop QueryWeaver from source.

Prerequisites

• Python 3.12+
• uv (Python package manager)
• A FalkorDB instance (local or remote)
• Node.js and npm (for the React frontend)

Install and configure

Quickstart (recommended for development):

# Clone the repogit clone https://github.com/FalkorDB/QueryWeaver.gitcd QueryWeaver
# Install dependencies (backend + frontend) and start the dev servermake installmake run-dev

If you prefer to set up manually or need a custom environment, use uv:

# Install Python (backend) and frontend dependenciesuv sync
# Create a local environment filecp .env.example .env# Edit .env with your values (set APP_ENV=development for local development)

Run the app locally

uv run uvicorn api.index:app --host 0.0.0.0 --port 5000 --reload

The server will be available at http://localhost:5000

Alternatively, the repository provides Make targets for running the app:

make run-dev   # development server (reload, debug-friendly)make run-prod  # production mode (ensure frontend build if needed)

Frontend build (when needed)

The frontend is a modern React + Vite app in app/. Build before production runs or after frontend changes:

make install       # installs backend and frontend depsmake build-prod    # builds the frontend into app/dist/
# or manuallycd appnpm cinpm run build

OAuth configuration

QueryWeaver supports Google and GitHub OAuth. Create OAuth credentials for each provider and paste the client IDs/secrets into your .env file.

• Google: set authorized origin and callback http://localhost:5000/login/google/authorized
• GitHub: set homepage and callback http://localhost:5000/login/github/authorized

Environment-specific OAuth settings

For production/staging deployments, session cookies are HTTPS-only by default. Only an APP_ENV that reads as development once trimmed and lower-cased turns that off, so a deployment that forgets the variable still gets secure cookies. Set APP_ENV=development for plain-HTTP local runs, otherwise the browser drops the cookie and you get OAuth CSRF state mismatch errors.

The signed session cookie is the browser's only credential. An api_token is never written to a browser cookie - a bearer token in a cookie sits on disk in clear text for its whole lifetime - so programmatic clients fetch one from the tokens API instead. Sessions issued before this change keep working: the legacy api_token cookie is still accepted, just no longer handed out.

# For production/staging (HTTPS-only session cookies - also the default)APP_ENV=production
# For development (allows HTTP session cookies)APP_ENV=development

Important: If you're getting "mismatching_state: CSRF Warning!" errors on a plain-HTTP environment, ensure APP_ENV is set to development.

AI/LLM configuration

QueryWeaver supports multiple AI providers. Set one API key and QueryWeaver auto-detects which provider to use.

Priority order: Ollama > OpenAI > Gemini > Anthropic > Cohere > Azure (default)

* Anthropic has no native embeddings. You must set VOYAGE_API_KEY or EMBEDDING_MODEL for embeddings, otherwise startup will fail with an error.

Optional: Override default models

COMPLETION_MODEL=gemini/gemini-3-pro-previewEMBEDDING_MODEL=gemini/gemini-embedding-001

Both must match your API key's provider.

To change only one provider's defaults, set <PROVIDER>_MODEL / <PROVIDER>_EMBEDDING_MODEL to a bare model name (e.g. AZURE_MODEL=my-gpt-deployment). COMPLETION_MODEL / EMBEDDING_MODEL take precedence over these. Anthropic has no ANTHROPIC_EMBEDDING_MODEL; its embeddings come from VOYAGE_API_KEY (fixed voyage/voyage-3) or EMBEDDING_MODEL.

Docker examples with AI configuration

Using OpenAI:

docker run -p 5000:5000 -it \  -e FASTAPI_SECRET_KEY=your_secret_key \  -e OPENAI_API_KEY=your_openai_api_key \  falkordb/queryweaver

Using Google Gemini:

docker run -p 5000:5000 -it \  -e FASTAPI_SECRET_KEY=your_secret_key \  -e GEMINI_API_KEY=your_gemini_api_key \  falkordb/queryweaver

Using Anthropic:

docker run -p 5000:5000 -it \  -e FASTAPI_SECRET_KEY=your_secret_key \  -e ANTHROPIC_API_KEY=your_anthropic_api_key \  falkordb/queryweaver

Using Azure OpenAI:

docker run -p 5000:5000 -it \  -e FASTAPI_SECRET_KEY=your_secret_key \  -e AZURE_API_KEY=your_azure_api_key \  -e AZURE_API_BASE=https://your-resource.openai.azure.com/ \  -e AZURE_API_VERSION=2025-03-01-preview \  falkordb/queryweaver

Testing

Quick note: many tests require FalkorDB to be available. Use the included helper to run a test DB in Docker if needed.

Prerequisites

• Install dev dependencies: uv sync
• Start FalkorDB (see make docker-falkordb)
• Install Playwright browsers: uv run playwright install

Quick commands

Recommended: prepare the development/test environment using the Make helper (installs dependencies and Playwright browsers):

# Prepare development/test environment (installs deps and Playwright browsers)make setup-dev

Alternatively, you can run the E2E-specific setup script and then run tests manually:

# Prepare E2E test environment (installs browsers and other setup)./setup_e2e_tests.sh
# Run all testsmake test
# Run unit tests only (faster)make test-unit
# Run E2E tests (headless)make test-e2e
# Run E2E tests with a visible browser for debuggingmake test-e2e-headed

Test types

• Unit tests: focus on individual modules and utilities. Run with make test-unit or uv run python -m pytest tests/ -k "not e2e".
• End-to-end (E2E) tests: run via Playwright and exercise UI flows, OAuth, file uploads, schema processing, chat queries, and API endpoints. Use make test-e2e.

See tests/e2e/README.md for full E2E test instructions.

CI/CD

GitHub Actions run unit and E2E tests on pushes and pull requests. Failures capture screenshots and artifacts for debugging.

Troubleshooting

• FalkorDB connection issues: start the DB helper make docker-falkordb or check network/host settings.
• Playwright/browser failures: install browsers with uv run playwright install and ensure system deps are present.
• Missing environment variables: copy .env.example and fill required values.
• OAuth "mismatching_state: CSRF Warning!" errors: Session cookies are HTTPS-only unless APP_ENV reads as development once trimmed and lower-cased. Set APP_ENV=development for plain-HTTP environments; use production or staging (or leave the variable out entirely) for HTTPS deployments.

Project layout (high level)

• api/ – FastAPI backend
• app/ – React + Vite frontend
• tests/ – unit and E2E tests

License

Licensed under the GNU Affero General Public License (AGPL). See LICENSE.

Copyright FalkorDB Ltd. 2025