Connectors

Braids ships with 32 built-in connectors covering payments, e-commerce, developer tools, communication, and more.

Built-in Connectors#

AI

Authentication

Cloud Infrastructure

Communication

CRM

Databases

Developer Tools

E-Commerce

Email

Infrastructure

Observability

Media

Payments

Project Management

Scheduling

Testing

Authentication Types#

bearer

Sends Authorization: Bearer <token>. This is the most common authentication method.

connectors:
  stripe:
    type: stripe
    config:
      api_key: ${STRIPE_API_KEY}

header

Custom header name, optionally with a prefix.

connectors:
  shopify:
    type: shopify
    config:
      store: mystore
      access_token: ${SHOPIFY_TOKEN}
# Sends: X-Shopify-Access-Token: <token>

Discord example with prefix:

connectors:
  discord:
    type: discord
    config:
      token: ${DISCORD_BOT_TOKEN}
# Sends: Authorization: Bot <token>

basic

HTTP Basic Auth (username:password base64 encoded).

connectors:
  jira:
    type: jira
    config:
      domain: mycompany
      email: ${JIRA_EMAIL}
      api_token: ${JIRA_TOKEN}

query_param

Token sent as a URL query parameter.

# Token appended to request URL as ?key=<value>

none

No authentication required.

connectors:
  jsonplaceholder:
    type: jsonplaceholder

extra_headers

Additional headers for multi-part authentication (e.g., Plaid).

connectors:
  plaid:
    type: plaid
    config:
      client_id: ${PLAID_CLIENT_ID}
      secret: ${PLAID_SECRET}
# Sends: PLAID-CLIENT-ID + PLAID-SECRET headers

Pagination Types#

cursor

Cursor-based pagination. Uses a cursor field from the response to fetch the next page. Used by: Stripe, Slack, HubSpot, Square.

link_header

RFC 5988 Link header. Follows rel="next" URL from the response Link header. Used by: GitHub, GitLab, Shopify, Sentry, DigitalOcean.

offset

Offset/limit pagination. Increments offset by limit each page. Used by: Jira, Asana, PagerDuty, Spotify.

page

Page number pagination. Increments the page number each request. Used by: Cloudflare.

Custom Connectors#

Use type: path to define custom connectors from a local directory. Each custom connector lives in its own folder with a connector.yaml file.

Directory structure:

connectors/
  my-api/
    connector.yaml

The connector.yaml file defines the base URL, authentication, and available resources:

name: my-api
description: My custom API
base_url: https://api.example.com
auth:
  type: bearer
  token_field: api_key
resources:
  users:
    path: /v1/users
    method: GET
    data_field: data
  user:
    path: /v1/users/{id}
    method: GET

Then reference the custom connector in your braids.yaml:

connectors:
  my_api:
    type: path
    path: ./connectors/my-api
    config:
      api_key: ${MY_API_KEY}

OpenAPI Import#

Connectors can auto-discover resources from OpenAPI specs. When a connector has an openapi_url defined, running braids connectors update downloads and caches the spec locally. Resources are auto-discovered from the spec — every path and method combination becomes an available resource.

Explicit resource definitions in your connector.yaml always override auto-discovered ones, so you can customize individual endpoints while still benefiting from automatic discovery for the rest.

GraphQL Connectors#

GraphQL connectors (like shopify-storefront) send GraphQL queries instead of REST requests. Configure them the same way as REST connectors:

connectors:
  storefront:
    type: shopify-storefront
    config:
      store: mystore
      storefront_token: ${STOREFRONT_TOKEN}

In endpoints, use query and variables fields. The connector sends the GraphQL query and extracts the data using the data_field path:

endpoints:
  /products:
    schema: products
    sources:
      - connector: storefront
        resource: products
        mapping:
          id: id
          title: title

The data_field extraction navigates nested GraphQL response structures automatically. For example, a response at data.products.edges[].node is flattened into a simple array of product objects before mapping is applied.