Skip to main content
Agents are AI-powered assistants that can use tools to solve open-ended tasks. Unlike workflows that follow predetermined paths, agents make intelligent decisions about which tools to use and when, adapting their approach based on the task at hand.

​
What Are Agents?

Think of agents as intelligent assistants that can orchestrate your workflows. You give them a goal, provide them with tools (integrations and workflows), and they figure out how to accomplish the task by deciding which tools to use and when. Key Characteristics:
  • Adaptive: Different approaches for different situations
  • Tool-driven: Use integrations and workflows as needed
  • Conversational: Interactive back-and-forth discussions
  • Context-aware: Consider your instructions and conversation history
Agents vs. Quarterbacks: Agents are like quarterbacks who read the situation and call the right plays (workflows). Workflows are the plays themselves—reliable, repeatable sequences that execute consistently.

​
Agents vs. Workflows

Understanding when to use agents versus workflows is critical for success:
AspectWorkflowsAgents
StructureDefined start, steps, and endOpen-ended task completion
ExecutionSame path every timeAdapts based on context
ReliabilityHigh (100% predictable)Medium (AI-driven decisions)
CostLow (fixed per run)Higher (based on AI usage)
SpeedFast (direct execution)Slower (reasoning required)
FlexibilityLow (rigid structure)High (dynamic approach)
Best ForRepetitive, high-volume tasksAd-hoc, decision-making tasks
  • Use Workflows When
  • Use Agents When

​
Workflows Excel At:

  • Defined processes with clear inputs, steps, and outputs
  • High-frequency operations (multiple times daily)
  • Mission-critical tasks requiring reliability
  • Predictable costs and execution times
  • Event-based triggers (new row, form submission, etc.)

​
Examples:

  • Data enrichment pipelines
  • Scheduled reports and notifications
  • Email automation sequences
  • Lead scoring systems
  • Batch processing operations

Start with Workflows

Build reliable automation “plays” that handle specific tasks consistently and efficiently

Graduate to Agents

Create agents that orchestrate when and how to use those workflows based on context

​
How Agents Work

Agents operate through a framework of Tools, Instructions, and Reasoning:

​
1. Tools (What Agents Can Use)

Agents are equipped with tools to accomplish tasks:
Direct connections to external services:
  • Gmail: Read, search, and send emails
  • Salesforce: Query records and update data
  • Notion: Search documentation and databases
  • Zendesk: Retrieve and manage support tickets
  • Google Calendar: Check availability and schedule meetings
  • And many more
Agents use the personal default credentials of whoever is running them (unless workspace credentials are configured). You’ll be prompted to authenticate if needed.
Your workflows become powerful tools for agents:
  • Agents see the workflow name and description
  • Understand expected inputs and outputs
  • Call workflows when appropriate
  • Process results to inform next steps
Example: A workflow that queries BigQuery, pulls Salesforce data, and combines them becomes a single “Get User Profile” tool for your agent.
Connect your own MCP server for services not available natively:
  • Internal company APIs
  • Custom data sources
  • Specialized tools
Learn more in the MCP documentation.
Agent tools including integrations and workflows

​
2. Instructions (How Agents Behave)

The system prompt defines your agent’s personality, behavior, and decision-making:

Define a Role

Good: “You’re an executive assistant for a CFO of a Fortune 500 company.”Bad: “You help with tasks.”A specific role sets tone, expertise level, and communication style automatically.

Set Tool Usage Rules

Explain when and how to use specific tools:
When the user provides a ticket number:
1. Retrieve ticket details and ask for confirmation
2. Once confirmed, run the "Get User Profile" workflow
3. Read the discount policy from Notion
4. Assess eligibility and explain your reasoning

Establish Confirmation Rules

Define when the agent should ask before acting:“Always ask for confirmation before:
  • Sending emails
  • Deleting data
  • Making changes to external systems”

Specify Response Style

Control tone, length, and format:“Always respond professionally. Keep responses under 100 words. Use bullet points for lists.”
Agent system prompt example

​
3. Reasoning (How Agents Think)

When you assign a task, the agent:
  1. Analyzes the request and available tools
  2. Decides which tools to use and in what order
  3. Executes tool calls and adapts based on results
  4. Asks for confirmation when needed (based on instructions)
  5. Explains its reasoning step-by-step
Chat with Agent

​
Credentials & Authentication

​
How Agents Use Credentials

When an agent uses integrations and workflows, it needs credentials to access external services. Understanding how credentials work is essential for secure and effective agent deployment.
Key principle: Agents always use the credentials of the person running the agent, not the agent creator’s credentials. If you run the agent, it uses your credentials. If a teammate runs it, it uses their credentials.
Workspace agents are created in shared workspaces and have two key differences from personal agents:1. Access Control:
  • Only members of the workspace can use the agent
  • Non-members will receive an “access denied” message
2. Credential Behavior:
  • If an MCP integration or workflow is set to use “workspace default” credentials, those workspace credentials are used instead of personal credentials
  • Otherwise, the personal default credentials of whoever is running the agent are used
Best for:
  • Team collaboration requiring shared credentials
  • Controlled access to specific workspace members
Learn more about workspaces in the Organizations and Workspaces documentation.
Before using an agent with integrations:
  1. Visit your credentials page
  2. Authenticate with required services (Gmail, Salesforce, etc.)
  3. Set credentials as your personal default
What happens if credentials are missing:
  • The agent will notify you about missing authentication
  • You’ll receive a link to the credentials page
  • After authenticating, return and retry your request
Agents will automatically detect missing credentials and guide you through the authentication process.

​
Data Privacy & Security

Your Data Stays Private

With personal agents and personal credentials, only your authenticated accounts are accessed. Other users cannot see your data through the agent.

Controlled Access

Admin security controls and user roles apply to agents just like workflows. You can only access what you’re authorized to access.

​
Creating Your First Agent

1

Add Tools

Select integrations and workflows your agent needs:Start Simple: Begin with 2-3 tools and add more as needed. Too many tools can overwhelm the agent and make behavior less predictable.
If you’re using workflows as tools, use descriptive names like “Get User Activity and Salesforce Status” instead of “User Workflow” so agents understand when to use them.
2

Limit Tool Capabilities (Recommended)

For MCP integrations, you can restrict which specific tools the agent can access. This makes your agent more reliable and prevents unintended actions.How to limit tools:
Click on integration to configure tools
  1. Click on any MCP integration you’ve added (Gmail, Salesforce, Slack, etc.)
  2. Toggle off specific tools you don’t want the agent to use
Limit available tools for integration
Example - Gmail Integration:
  • ✅ Enable: Search emails, Read emails
  • ❌ Disable: Send email, Delete email
Example - Salesforce Integration:
  • ✅ Enable: Get account, Get opportunity, Search records
  • ❌ Disable: Delete record, Update record
Example - Slack Integration:
  • ✅ Enable: Read messages, Search channels
  • ❌ Disable: Send message, Create channel
Why this matters:
  • Prevents destructive actions (deleting, sending)
  • Makes agent behavior more predictable
  • Reduces risk of unintended operations
  • Improves agent reliability by limiting decision space
Pro tip: Add this to your system prompt: “If a user asks you to perform an action that’s disabled (like sending an email), explicitly tell them that capability is not available for this agent.”
3

Write Instructions

Create clear, specific instructions:
You're a support operations assistant helping evaluate discount eligibility.

When given a Zendesk ticket number:
1. Retrieve ticket details and display them
2. Ask for confirmation before proceeding
3. Run "Get User Profile" workflow
4. Read discount criteria from Notion page [URL]
5. Evaluate eligibility with clear reasoning

Always respond professionally. If uncertain, ask for clarification.
Your agent is brilliant but new to your business. Be explicit about your preferences and expectations.
4

Test Thoroughly

Use the built-in chat interface to test:
  • Start Simple: Test basic functionality with straightforward requests
  • Find Edge Cases: Try unexpected inputs and ambiguous requests
  • Refine Instructions: When mistakes occur, ask the agent: “What could I add to your instructions to help you handle this correctly next time?”
  • Document Patterns: Keep notes on successful approaches
Agent builder interface

​
Understanding Credit Costs

Agents consume credits based on AI model usage, workflow executions, and integration operations.

​
How Credits Work

Credits are charged per AI interaction. The cost depends on:
  • Message length: Longer messages consume more credits
  • Model selected: Different AI models have different credit costs
  • Conversation history: Each message includes previous context
  • Tools available: More tools slightly increase base cost
Each time your agent reads your message, thinks about what to do, and responds, it consumes credits based on the AI model you’ve selected.

​
Model Pricing Overview

  • Budget Models
  • Advanced Models
  • Expert Models
Best for simple tasks and high-volume usage
Model~Cost per MessageBest For
GPT-4.1 Mini2-3 creditsSimple tasks, quick responses
GPT-5 Mini2-3 creditsGeneral queries
Claude Haiku 4.52-4 creditsFast interactions
Note: Credit costs shown are approximate per message. Actual costs vary based on message length, conversation history, number of tools available, and whether the agent executes workflows during the interaction.

​
What Affects Credit Costs

Longer inputs and outputs consume more tokens:
  • Short message (50 words): ~500 tokens
  • Medium message (200 words): ~2,000 tokens
  • Long message (500 words): ~5,000 tokens
Each message includes previous conversation context:
  • 1st message: Base cost
  • 5th message: ~20% higher (accumulated history)
  • 10th message: ~40% higher
Tip: Start new conversations for unrelated topics.
More tools = larger system prompt:
  • 2-3 tools: Base cost
  • 5-10 tools: +10-20% cost
  • 15+ tools: +30-40% cost
Workflow calls trigger multiple AI steps:Example: Agent runs a workflow
  1. Decide to call workflow: ~5 credits
  2. Process workflow results: ~5 credits
  3. Generate final response: ~5 credits
Total: ~15 credits + workflow execution cost

​
Workflow and Integration Costs

When agents call workflows or integrations, additional costs apply: Workflow Execution Costs:
Node TypeCredit CostExamples
Free Nodes0 creditsInput/Output, Filter, Router, Most integrations
Low Cost2-3 creditsAsk AI (simple), Run Code, Custom Operators
Medium Cost10-30 creditsAI with large prompts, AI Vision
High Cost10-60+ creditsData enrichment, Premium APIs
Integration Costs:
  • Free: Google Workspace, Slack, Airtable, Notion, Salesforce, HubSpot, 150+ more
  • Paid: Data enrichment services (eg. Hunter.io ~10 credits, ZoomInfo ~60 credits)
Total Cost Formula: 
Total = AI Model Cost (reasoning) + Workflow Cost (if applicable) + Integration Cost (if applicable)

​
Optimizing Credit Usage

Choose Appropriate Models

Use budget models for simple tasks. Reserve expert models for complex analysis that truly requires advanced reasoning.

Keep Conversations Focused

Start new conversations for different topics. Long threads accumulate context costs with each message.

Write Clear Prompts

Specific prompts get better results faster, reducing back-and-forth messages and total credit consumption.

Limit Tool Count

Start with 2-3 essential tools. Each additional tool increases the system prompt size and base cost.

​
Tracking Credit Usage

Monitor your credit consumption across multiple views:
Agent credit cost
  • Real-time Display: Credits shown next to each agent response as it streams
  • Conversation History: View total credits per conversation thread
  • Workflow Runs: See individual workflow execution costs when agents call them
  • Account Dashboard: Track total credits used across all agents and timeframes

​
Using Workflows as Agent Tools

The most powerful pattern is using workflows as tools for agents. This transforms you from manual orchestrator to strategic designer.

Before Agents

You are the orchestrator:
  • Monitor conditions manually
  • Decide when to run workflows
  • String workflows together yourself
  • Handle exceptions and edge cases

With Agents

Agent orchestrates automatically:
  • Monitors and responds to requests
  • Decides which workflows to use
  • Chains workflows intelligently
  • Adapts to results and conditions

​
Designing Workflows for Agents

When building workflows that agents will use as tools, follow these practices:

Use Input and Output Nodes

Critical: Agents identify workflow parameters through Input and Output nodes.Good Practice:
Input Node: "company_name" (text)
Input Node: "date_range" (text)
[Your workflow logic here]
Output Node: "enriched_data" (object)
Output Node: "success" (boolean)
Why: The agent reads these nodes to understand what data the workflow expects and what it returns. Without clear inputs/outputs, the agent won’t know how to use the workflow properly.

Use Descriptive Names

Workflow names should clearly indicate their purpose:✅ Good: “Get User Activity and Salesforce Status”
✅ Good: “Enrich Lead from LinkedIn Profile”
✅ Good: “Check Discount Eligibility from Ticket”❌ Bad: “User Workflow”
❌ Bad: “Flow 1”
❌ Bad: “Data Thing”

Add Clear Descriptions

Write descriptions that help agents understand:
  • What the workflow does
  • When it should be used
  • What inputs it expects
  • What outputs it provides
Example: “Takes a LinkedIn profile URL and returns enriched company data including size, industry, recent funding, and key contacts. Use when researching new leads or companies.”

Keep Workflows Focused

Each workflow should do one thing well:
  • ✅ “Enrich Contact from Email”
  • ✅ “Send Slack Notification”
  • ❌ “Enrich Contact and Send Notification and Update CRM”
Let the agent orchestrate multiple focused workflows rather than building mega-workflows.

​
Best Practices

Begin with 2-3 tools and straightforward instructions. Test thoroughly before adding more capabilities.Anti-pattern: Creating an agent with 15 tools and a 2000-word system prompt immediately.Better approach: Start with core functionality, observe how the agent behaves, then incrementally add tools and refine instructions based on real usage.
Agents improve over time as you refine instructions based on real interactions:
  • Review conversation history regularly for patterns
  • When agents make mistakes, ask: “What could I add to your instructions to prevent this?”
  • Document edge cases and add explicit handling rules
  • Celebrate successful patterns and codify them in instructions
  • Share learnings across team members using the agent
When creating workflows that agents will call:
  • Always use Input and Output nodes so agents understand parameters
  • Keep workflows focused on single responsibilities
  • Use descriptive names that indicate purpose
  • Write clear descriptions explaining when to use the workflow
  • Test workflows independently before giving them to agents
Define what agents should NOT do to prevent unintended actions:
Never:
- Delete customer data or records
- Send emails without explicit user approval
- Make purchases or financial commitments
- Override manual decisions by team members
- Modify production data without confirmation
Be explicit about destructive actions requiring human oversight.
Track metrics to understand agent effectiveness:
  • Time saved: Hours saved per week through automation
  • Success rate: Tasks completed successfully vs requiring intervention
  • Tool usage: Which workflows/integrations are used most frequently
  • Credit efficiency: Cost per completed task or interaction
  • User satisfaction: Team adoption and feedback
Use insights to optimize instructions and expand capabilities strategically.

​
Troubleshooting

Symptoms: Agent calls inappropriate tools or skips necessary stepsSolutions:
  • Make tool names more descriptive and explicit
  • Add specific “When to use” guidance in system prompt
  • Reduce number of similar tools that might confuse the agent
  • Provide examples: “When user asks X, use tool Y”
  • Check if workflow descriptions clearly explain their purpose
Symptoms: Agent fails to call workflows correctly or misinterprets resultsSolutions:
  • Ensure workflows have Input and Output nodes defined
  • Use clear, descriptive names for input/output fields
  • Add workflow descriptions explaining parameters
  • Test workflow independently to verify inputs/outputs work
  • Simplify complex workflows into smaller, focused ones
Symptoms: Agent reports missing credentials or authentication failuresSolutions:
  • Visit your credentials page
  • Authenticate with the required service
  • Set it as your personal default credential
  • For workspace agents using workspace credentials, contact your workspace admin
  • Return to the agent and retry your request
Symptoms: Agent seeks approval for routine operations, slowing down interactionsSolutions:
  • Explicitly list operations that don’t need confirmation
  • Add “proceed automatically when…” clauses for common scenarios
  • Specify conditions under which confirmation IS required
  • Review and tighten confirmation rules in system prompt
Symptoms: Agent takes sensitive actions without required approvalSolutions:
  • Add explicit “always ask before…” rules for destructive actions
  • List all sensitive operations requiring human approval
  • Include examples of when to ask vs when to proceed
  • Test with dry-run scenarios before giving real access
Symptoms: Responses too verbose, too terse, wrong tone, or poor formattingSolutions:
  • Update role definition to set appropriate tone
  • Add response format examples in instructions
  • Specify length constraints: “Keep responses under 100 words”
  • Request specific formatting: “Use bullet points for lists”
  • Provide example responses that match desired style
Symptoms: Agent stops mid-task, declares tasks impossible, or gets stuckSolutions:
  • Verify all required tools are properly connected
  • Check authentication for all integrations (agents will prompt if needed)
  • Review error logs in conversation history for specific failures
  • Simplify initial instructions and build up complexity gradually
  • Test individual tools work before expecting agent to orchestrate them
  • Check if workflows have errors that prevent agent from calling them
Symptoms: Agent consuming more credits than expectedSolutions:
  • Switch to a less expensive model for simple tasks
  • Start new conversations instead of continuing long threads
  • Reduce number of tools to decrease system prompt size
  • Optimize workflows to use fewer AI nodes
  • Write clearer prompts to reduce back-and-forth messages
  • Review if agent is making unnecessary tool calls

​
Where to Use Agents

Agents can be used in multiple ways across your workflow ecosystem:
  • Chat Interface
  • Slack Integration
Interact with agents directly in Gumloop for ad-hoc tasks, testing, and development.Accessing the chat interface:
  1. Go to your Agents page
  2. Click on the share button on the top
  3. Copy the chat interface link and use it to chat with the agent in your browser
Agent chat interface location
Sharing the chat interface: You can share a direct link to chat with your agent so others can use it directly. When someone uses your shared chat link, the agent will use their personal credentials, not yours.

​
Next Steps

Browse Agent Templates

Explore pre-built agents for inspiration and quick starts

Using Agents in Slack

Learn how to deploy agents to Slack channels

Agents Course

Watch video tutorials on building and optimizing agents
⌘I