Collections

Collections store reusable requests for HTTP, GraphQL, gRPC, and WebSocket. A saved request contains the protocol target, payload, headers or metadata, authorization, scripts, response history settings, and documentation.

Use Collections when the same request must be sent again, documented, or shared with your team.

Editor layout

The Collections workspace is split into these areas:

AreaPurpose
Collection treeStores collections, folders, and requests. Folders can define inherited auth, variables, and scripts.
Request headerContains the request name, protocol or method selector, target URL or endpoint, Save, Send, environment selector, and request tools.
Request tabsEdit the selected request. Tabs change by protocol.
Response panelShows the latest response, status, duration, response headers, and response body for protocols with a direct response view.
Variables panelShows every {{variable}} used by the request, where it is defined, and whether it is missing.
Diagnostics drawerShows console logs, script output, issues, network history, and runtime details.

Common tabs

Every protocol starts with a Docs tab and ends with Scripts and Settings.

TabUsed for
DocsMarkdown notes for the request: behavior, input examples, expected response, and ownership notes.
AuthorizationRequest-level auth. A request can inherit auth from its collection or folder, or override it.
HeadersHTTP-style headers for HTTP, GraphQL, and WebSocket.
MetadatagRPC metadata key-value pairs.
BodyHTTP request body.
QueryGraphQL query or mutation text.
VariablesGraphQL variables JSON.
MessagegRPC message editor.
MessagesWebSocket message editor.
SchemaGraphQL schema configuration.
Service definitiongRPC proto or reflection configuration.
ConnectionProtocol-specific connection settings.
EventsWebSocket message event handling.
ScriptsPre-request and post-response automation for the request.
SettingsProtocol options such as TLS validation, redirects, timeouts, response history, reflection, or message limits.

Protocol tabs

ProtocolTabs
HTTPDocs, Params, Authorization, Headers, Body, Scripts, Settings
GraphQLDocs, Query, Variables, Authorization, Headers, Schema, Scripts, Settings
gRPCDocs, Message, Authorization, Metadata, Service definition, Scripts, Settings
WebSocketDocs, Messages, Authorization, Headers, Connection, Events, Scripts, Settings

HTTP requests

HTTP requests use a method, URL, optional query params, headers, auth, and body.

TabWhat to configure
ParamsQuery string rows. Enabled rows with a key are appended to the URL.
AuthorizationInherited auth or request-specific auth. Use this for bearer tokens, basic auth, API keys, and other supported auth modes.
HeadersRequest headers. The default Content-Type is application/json.
BodyRaw JSON or text, form data, URL encoded data, binary data, or no body depending on body mode.
ScriptsPre-request runs before Send. Post-response runs after a response.
SettingsRedirect handling, TLS certificate validation, variable resolution before send, and response history behavior.

Example HTTP target:

GET {{host}}/api/internal/tokens?limit={{limit}}

GraphQL requests

GraphQL requests are saved as a query or mutation plus variables and HTTP-style request options.

TabWhat to configure
QueryGraphQL query or mutation text.
VariablesJSON object used as GraphQL variables. The tab label shows a count when variables are present.
AuthorizationAuth inherited from collection or folder, or request-specific auth.
HeadersHeaders for the GraphQL HTTP request.
SchemaSchema source, schema URL, or schema definition.
ScriptsBefore query can prepare headers and variables. After response can assert GraphQL response data or store values.
SettingsOperation name and GraphQL-specific execution settings.

Example:

query Account($id: ID!) {
  account(id: $id) {
    id
    balance
  }
}
{
  "id": "{{account_id}}"
}

gRPC requests

gRPC requests use service and method selection, metadata, a JSON message, and proto or reflection settings.

TabWhat to configure
MessageJSON request message.
AuthorizationAuth data that can be translated into request metadata.
MetadatagRPC metadata key-value rows.
Service definitionReflection or service definition source, service name, and method name.
ScriptsBefore invoke, On message, and After response.
SettingsTLS or plaintext mode, TLS server name, reflection timeout, response message size, default value emission, and call timeout.

Use gRPC requests for service-to-service APIs where a proto service and method define the contract.

WebSocket requests

WebSocket requests model a connection plus one or more messages.

TabWhat to configure
MessagesText or JSON message payloads to send after connection.
AuthorizationHandshake auth.
HeadersHandshake headers.
ConnectionWebSocket URL, protocols, timeout, reconnect behavior, and message type.
EventsMessage lifecycle and received message handling.
ScriptsBefore connect can prepare URL, headers, and variables. On message can inspect incoming messages.
SettingsWebSocket execution options.

Use WebSocket requests for realtime APIs, subscriptions, and streaming message checks.

Variables and scripts

Requests use {{variable}} templates resolved from collection and folder variables, the selected environment, and script-created values. See Environments and Scripts.

Share docs

Share docs on a collection's Overview publishes the collection's documentation as a read-only web page. Anyone with the link can read it without an account, so you can send API docs to a partner, a customer, or a teammate who doesn't use Requestarium.

The page lists every request in tree order, grouped by folder, with its method, URL, docs (Markdown), headers, and, if you keep Include request bodies as examples on, its body. Header values that look like credentials (Authorization, Cookie, API keys, tokens) are masked unless they are a {{variable}}; check bodies for secrets before you share.

  • The page is a snapshot. After you change the collection, open Shared docs and click Update page: the content is replaced and the link stays the same.
  • Copy link copies the address. The link is long and random, and the page asks search engines not to index it.
  • Turn off link stops the page at once. Sharing again makes a new link, so the old one never works again.

Publishing and turning the link off need write access to the collection. Sharing docs is available on every plan.