Docs Navigation

Blocks

Agent

Build an agent

Create powerful AI agents using any LLM provider with customizable system prompts and tool integrations.

Back to Catalog

Agentsagent

Block Preview

Agent

Build an agent

Usage

Add the block to your workflow and connect it to the upstream step.
Configure any required credentials or tokens in the inputs.
Fill in required inputs and optional parameters for the run.
Run a test execution, inspect outputs, and iterate before deploying.
Deploy the agent block with monitoring enabled in production.

Inputs (UI)

System Prompt

long-input

Placeholder: You are a helpful AI assistant. Be concise and professional.

Layout: full

User Prompt

long-input

Placeholder: Enter user message here. Use %input% for variables.

Layout: full

Model

dropdown

Layout: half

Options: Dynamic options

Temperature

slider

Layout: half

Condition: model = ["gpt-4o","gpt-4o-mini","gpt-4-turbo","gpt-4","gpt-3.5-turbo","o3-mini","gpt-5-mini","gpt-5","o1-mini","o1-preview","gemini-2.5-pro-preview-05-06","gemini-2.5-flash-preview-05-20","gemini-2.0-flash","gemini-2.0-flash-lite","deepseek-chat","deepseek-reasoner","llama-4-scout-17b-16e-instruct","llama-4-maverick-17b-128e-instruct","llama-3.3-70b-versatile","qwen-qwq-32b","deepseek-r1-distill-llama-70b"]

Temperature

slider

Layout: half

Condition: model = ["claude-opus-4-5-20251022","claude-sonnet-4-20250514","claude-opus-4-20250514","claude-haiku-4-5-20251001","claude-3-7-sonnet-20250219","grok-3-latest","grok-3-fast-latest"]

API Key

short-input

Placeholder: sk-... or use environment variable like $OPENAI_API_KEY

Layout: full

Condition: model = ["qwen2.5:0.5b","qwen2.5:1.5b","qwen2.5:3b","llama3.2:1b","llama3.2:3b","phi3.5:3.8b"]

Tools

tool-input

Layout: full

Response Format (Optional)

code

Placeholder: Leave empty for natural language response, or define JSON schema: { "type": "object", "properties": { "answer": {"type": "string"} }, "additionalProperties": false }

Layout: full

Knowledge Sources

knowledge-source-input

Layout: full

Max Search Results

slider

Layout: half

Condition: knowledgeSources = ""

Similarity Threshold

slider

Layout: half

Condition: knowledgeSources = ""

Enable Memory

switch

Layout: half

Memory Limit

slider

Layout: half

Condition: memoryEnabled = true

Importance Threshold

slider

Layout: half

Condition: memoryEnabled = true

Memory Tags

short-input

Placeholder: customer-support, billing, technical

Layout: half

Condition: memoryEnabled = true

Inputs (API)

systemPrompt

string

Optional

context

string

Optional

model

string

Required

apiKey

string

Required

responseFormat

json

Optional

Define the expected response format using JSON Schema. If not provided, returns plain text content.

temperature

number

Optional

tools

json

Optional

knowledgeSources

string

Optional

searchMaxResults

number

Optional

similarityThreshold

number

Optional

memoryEnabled

boolean

Optional

memoryLimit

number

Optional

memoryImportance

number

Optional

memoryTags

string

Optional

Outputs

Primary response type:

{
  "content": "string",
  "model": "string",
  "tokens": "json",
  "toolCalls": "json"
}

Conditional output based on: responseFormat

When empty: {"content":"string","model":"string","tokens":"json","toolCalls":"json"} | When filled: "json"

Tool Access

openai_chatanthropic_chatgoogle_chatxai_chatdeepseek_chatdeepseek_reasoner