Amazon Bedrock

Invoke foundation models on AWS Bedrock including Claude, Llama, and Titan.

AI and MLCredentials form5 actionsAll integrations

On this page

Connect
Call it from a workflow
Actions (5)
generate_content_from_image
generate_embeddings
generate_image
send_prompt
custom_api_call
Details

Connect

Auth type: Credentials form (custom_auth). Fill in the fields the provider needs (listed below). They are stored encrypted in the vault.

Field	Type	Description
`accessKeyId`required	text	Your AWS access key ID
`secretAccessKey`required	secret	Your AWS secret access key
`region`required	text	AWS region (e.g. us-east-1)

In the editor: Backend panel, Integrations tab, then Connect. This connects it for the project.
For every project in a workspace: Workspace settings, Integrations.
Owners and admins choose which integrations members may connect in the admin panel (Integrations tab), and can attach their own OAuth app there.

Call it from a workflow

Pass the action name and its input. The call resolves with { success, output, error? }; a missing connection resolves with success false and the reason. See ctx.integrations.invoke and ctx.integrations.call.

Native workflow

// .workflow.js (native ctx)
export async function run(ctx) {
  const res = await ctx.integrations.invoke("amazon_bedrock", "generate_content_from_image", {
    "model": "<model>",
    "image": "<image>",
    "prompt": "<prompt>"
  });
  if (!res.success) throw new Error(res.error);
  ctx.log.info("Amazon Bedrock answered", { output: res.output });
}

defineWorkflow module

// defineWorkflow module (@app/sdk ctx)
import { defineWorkflow } from "@app/sdk";

export default defineWorkflow({
  id: "amazon-bedrock-generate-content-from-image",
  name: "Amazon Bedrock: Generate Content from Image",
  description: "Calls Amazon Bedrock when a row is created.",
  trigger: { type: "entity_event", entity: "tickets", event: "created" },
  async run(ctx, event) {
    const res = await ctx.integrations.call("amazon_bedrock", {
      "action": "generate_content_from_image",
      "model": "<model>",
      "image": "<image>",
      "prompt": "<prompt>"
    }) as { success: boolean; error?: string };
    if (!res.success) throw new Error(res.error);
  },
});

From an app (connector proxy)

curl -X POST "$APP_ORIGIN/__appbricx/connector-proxy/amazon_bedrock/generate_content_from_image" \
  -H "Authorization: Bearer $APP_KEY" \
  -H "content-type: application/json" \
  -d '{"props":{"model":"<model>","image":"<image>","prompt":"<prompt>"}}'

Actions

generate_content_from_imagein catalog

Generate Content from Image. Ask a Bedrock model a question about an image.

Input	Type	Description
`model`required	id (picked from your account)	The foundation model to use. Must support image input.
`image`required	file	The image to analyze (PNG, JPEG, GIF, or WebP).
`prompt`required	long text	What do you want the model to tell you about the image?
`systemPrompt`optional	long text	Instructions that guide the model behavior.
`temperature`optional	number	Controls randomness. Lower values produce more deterministic output.
`maxTokens`optional	number	The maximum number of tokens to generate.

generate_embeddingsin catalog

Generate Embeddings. Generate vector embeddings from text using Amazon Titan Embed, Cohere Embed, or Amazon Nova Multimodal Embeddings models.

Input	Type	Description
`model`required	id (picked from your account)	The embedding model to use.
`inputText`required	long text	The text to generate embeddings for.
`embeddingPurpose`optional	one of "GENERIC_INDEX" \| "GENERIC_RETRIEVAL" \| "TEXT_RETRIEVAL" \| "CLASSIFICATION" \| "CLUSTERING"	Optimize embeddings for your use case. Only used by Nova Multimodal Embeddings.
`dimensions`optional	number	The number of dimensions for the output embedding vector. Titan Embed v2: 256, 512, 1024. Nova Multimodal: 256, 384, 1024, 3072.
`normalize`optional	boolean	Whether to normalize the output embedding vector. Supported by Titan Embed v2.

generate_imagein catalog

Generate Image. Generate an image from a text prompt using Amazon Titan Image Generator or Stability AI models.

Input	Type	Description
`model`required	id (picked from your account)	The image generation model to use.
`prompt`required	long text	A text description of the image you want to generate.
`negativePrompt`optional	long text	Describe what you do NOT want in the image. Helps refine the output.
`width`optional	number	Image width in pixels. Must be supported by the model.
`height`optional	number	Image height in pixels. Must be supported by the model.
`seed`optional	number	A seed for reproducible results. Use the same seed and prompt to get the same image.

send_promptin catalog

Ask Bedrock. Send a text prompt to an Amazon Bedrock model.

Input	Type	Description
`model`required	id (picked from your account)	The foundation model to use for generation.
`prompt`required	long text	Prompt
`systemPrompt`optional	long text	Instructions that guide the model behavior.
`temperature`optional	number	Controls randomness. Lower values produce more deterministic output.
`maxTokens`optional	number	The maximum number of tokens to generate.
`topP`optional	number	Nucleus sampling: the model considers tokens with top_p probability mass.
`stopSequences`optional	array	Sequences that will cause the model to stop generating. Up to 4 sequences.
`memoryKey`optional	text	A memory key that will keep the chat history shared across runs and flows. Keep it empty to leave the model without memory of previous messages.

custom_api_call

Custom API Call. Make a custom API call to any AWS Bedrock endpoint. Requests are automatically signed with AWS Signature V4.

Input	Type	Description
`service`required	one of "bedrock" \| "bedrock-runtime" \| "bedrock-agent" \| "bedrock-agent-runtime"	Select the AWS Bedrock service endpoint to call. Use "Bedrock" for control plane operations (e.g., ListFoundationModels) and "Bedrock Runtime" for inference operations (e.g., InvokeModel).
`method`required	one of "GET" \| "POST" \| "PATCH" \| "PUT" \| "DELETE" \| "HEAD"	Method
`path`required	text	The API path (e.g., /foundation-models, /model/{modelId}/invoke). See the AWS Bedrock API Reference for available endpoints.
`headers`optional	object	Additional headers to include. Authorization headers are automatically added via AWS Signature V4.
`queryParams`optional	object	Query parameters to include in the request URL.
`body`optional	JSON	JSON body for POST/PUT requests.
`failsafe`optional	boolean	No Error on Failure
`timeout`optional	number	Timeout (in seconds)

Details

Id: amazon_bedrock
Category: AI and ML
Tags: ai, aws, bedrock, llm, enterprise
Package: @activepieces/piece-amazon-bedrock@0.0.4
Tier: Built in