screwdriver-wrenchPipeline as a Tool

Pipeline/RAG as a Tool

Related page: Tool API Endpoint

The Standard RAG pipeline exposes two built-in tools that can be called directly via the Component API, without going through a full chat message flow. This is useful for integrations, agent workflows, or any scenario where you want retrieval or generation as a standalone operation.

Available Tools

Tool Name
Description
Subpage

standard-rag

Full RAG: retrieval + LLM answer generation

retrieval-pipeline

Retrieval only: fetch chunks from knowledge base

The tool name is used as part of the component_id in the endpoint:

POST /components/{chatbot_id}:{tool_name}/run

For a general overview of the Component API and how build_tools() works, see the Tool API Endpoint.

Authentication

All tools require a User API Key passed via the X-API-Key header.

Key format:

glc{version}-{user_id}:{hex_secret}

Example:

glc1-3cdda853-e800-471d-ac35-5fb5f1a0ae02:bc54c00fx8d69786aef8b03a14b17acdfef200bef8kf7aa0460bee805ec57e20

The calling user must have the SEND_MESSAGE permission and access to the target chatbot.

How to get your User API Key

Note: UI support for managing User API Keys is planned for a future release. For now, retrieve your key via these endpoints.

Step 1 — Log in to get an access token:

The response includes an access_token.

Step 2 — Retrieve your User API Key:

The response returns your api_key in plain text. Use this value in the X-API-Key header for all component tool calls.

Request Body

All tools share the same top-level request shape:

Field
Type
Required
Description

inputs

dict

Yes

Tool-specific inputs — see each tool's subpage for the full schema

config

dict

Yes

Runtime configuration overrides — see Runtime Config

model_name

string / null

No

LLM model to use. Falls back to chatbot default if omitted

Full Endpoint Example

Last updated