Skip to main content

#Trailhead Challenges1,138 人がディスカッション中

Hands-on challenges are the “secret sauce” of Trailhead. Before searching for solutions to superbadge challenges, review the Salesforce Certification Program Agreement and Policies. ** NOTE ** : If you were able to get a response that solved your issue, please mark it as the 'Best Answer' to help other Trailblazers. If the issue persists after 48 hours, create a Trailhead Help case at https://help.salesforce.com/s/support for further assistance.

Salesforce Developer Roadmap in the AI Era (2026–2027) 

 

Hi Trailblazers, 

 

With Agentforce, Data 360, and AI becoming part of Salesforce development, I believe developers need to strengthen their core technical skills while learning new AI capabilities.

Here is my suggested learning roadmap for beginners and working professionals. 

 

1. Salesforce Fundamentals

  • CRM basics, Objects, Fields, and Relationships
  • Security, OWD, Profiles, and Permission Sets
  • Flow Builder and Validation Rules

Salesforce Developer Beginner

 

2. SOQL & Apex Development

  • SOQL, SOSL, and Relationship Queries
  • Apex Classes, Triggers, and Collections
  • Bulkification, Governor Limits, and Test Classes

Apex Basics

 

3. Lightning Web Components (LWC)

  • HTML, CSS, and JavaScript
  • Components, Events, and Data Binding
  • LWC with Apex and Lightning Data Service

LWC Developer Guide

 

4. Salesforce Integrations

  • REST APIs, JSON, and OAuth 2.0
  • Named Credentials and Apex Callouts
  • Platform Events and Asynchronous Apex

Apex Integration Services

 

5. Salesforce DevOps

  • VS Code, Salesforce CLI, and Git
  • Sandboxes, Deployments, and Testing
  • CI/CD and DevOps Center

Salesforce CLI

 

6. Agentforce & AI Development

  • Agentforce Builder, Subagents, and Actions
  • Agent Script and Prompt Builder
  • Apex and Flow Actions for AI Agents
  • Agent Testing, Security, and Debugging

Agentforce Developer Guide

 

7. Data 360

  • Data Streams and Data Lake Objects
  • Data Mapping and Identity Resolution
  • Data Graphs, Retrieval, and Agentforce Grounding

Data 360 Learning

Data 360 Overview – SFDCINDIA

 

8. Real-World Projects

Build at least three practical projects:

  1. Student Management System – Admin, Apex, and LWC
  2. Customer Support Application – Flow, Integration, and Security
  3. AI Customer Support Agent – Agentforce, Actions, and Data 360

My Suggestion:

Don't jump directly into Agentforce without understanding Salesforce fundamentals.

Learn Salesforce → SOQL → Apex → LWC → Integration → DevOps → Agentforce → Data 360

AI can help developers write code faster, but understanding business requirements, architecture, security, debugging, and testing will remain essential.

My Key Point: The future belongs to developers who understand both Salesforce Development and AI.

What skills are you focusing on to stay relevant in the AI era? 

 

Venkat Yadav | SFDCINDIA

#Salesforce Developer #Salesforce Admin #Agentforce #Data Cloud #TrailblazerCommunity #Trailhead Challenges

0/9000

In the "Connect Flows to Agentforce" challenge within the "Become an Agentforce Innovator" module,  Agentforce is not working at all and an error message is impeding my progress, stating "Something went wrong. Refresh and try again." I've tried restarting the module and refreshing the page a couple of times, but nothing is working 

In Connect Flows to Agentforce challenge, Agentforce is not working

 

and it seems to be in continuous error. 

 

 

 

#Trailhead Challenges  #Trailhead  #Agentforce

2 件の回答
  1. 今日、4:39

    Hi @Aaron Alexander I checked your screenshot. The issue appears to be related to Agentforce Builder compilation, not necessarily your Flow configuration.

    I can see two errors:

    • "We couldn't compile the script. Try again later."
    • "Something went wrong. Refresh and try again."

    Please try these steps:

    1. Check Script View: Open Agent Script and look for syntax or configuration issues.
    2. Review Warnings: Check the 7 warnings shown in the Problems panel.
    3. Verify Flow Actions: Confirm the Flows are active and their input/output variables are configured correctly.
    4. Check Agent Settings: Make sure the agent and its subagents are configured correctly.
    5. Save and Retry: Save the draft, reopen Agentforce Builder, and test again.
    6. Try a New Version: If possible, create a new draft version and check whether the issue continues.

    Official Salesforce Resources:

    Since refreshing hasn't helped, focus on the compilation error first. If the problem continues with valid configurations, it could be a Salesforce platform-side issue.

    You can raise a Trailhead support case with the screenshot and exact error message. 

     

    Hope this helps! Venkat Yadav | SFDCINDIA

0/9000

For the tutorial Add a Flow as an Agent Action,  when I attempt to test the agent during the tutorial and the challenge, I receive the error "We couldn't compile the script. Try again later." I have tried connecting with a new developer account, however i'm still unable to complete this training.  

 

 

Error in Add a Flow as an Agent Action

 

image.png

 

 

 

image.png

 

 

 

 

 

#Trailhead Challenges

2 件の回答
  1. 今日、4:34

    Hi @Gibran Waldron I checked your screenshots. The error "We couldn't compile the script. Try again later" appears to be related to Agentforce Builder compilation, rather than the Flow itself.

    Please try these steps:

    1. Check Agent Script: Open Script View and look for any syntax or configuration issues.
    2. Verify Action Inputs: Make sure the Flow action inputs and outputs are mapped correctly, especially the required variables.
    3. Check Variables: Review any undefined variables or incorrect data types.
    4. Review Warnings: Your screenshots show three deprecated topic warnings. These may need updating, but they aren't necessarily causing the compilation failure.
    5. Save and Retry: Save the agent, refresh Agentforce Builder, and try compiling again.
    6. Test the Flow Separately: Run the Flow using Flow Debugger to confirm it works independently.

    Official Salesforce Resources:

    My Suggestion:

    Since you've already tried a new Developer Org, I would first check the Agent Script and action mappings. If everything looks correct and the error continues, it may be a platform-side compilation issue. 

     

    Hope this helps! Venkat Yadav | SFDCINDIA

0/9000

I'm currently working on the "Create Apex Agent Job" challenge as part of the "Agent customization via Apex" 

badge. 

 

Despite having completed all the required steps in my designated org (in two new different orgs requested), I keep running into a verification error. 

 

Here's what I've already checked:

 

  • I'm verifying the challenge against the correct org username: apexlegendbadge@sfdc.comand for the other org : test2@legend.apex
  • I've successfully created and configured both the Apex class and the custom action as instructed
  • The "Einstein Copilot" agent is set up and currently in Active status
  • I've tried clearing my browser cache and logging out and back in — the error still persists
  • I've tried in another browser and I have tried in an incognito window... error still here

 

Has anyone run into this recently? Any suggestions or workarounds would be greatly appreciated. Thanks in advance! 

 

#Trailhead Challenges  #Trailhead

0/9000

Hi Trailblazers, 

 

Salesforce has introduced Headless Experience Layer (HXL), an interesting addition for building interactive Agentforce experiences.

Instead of showing only text responses, Agentforce can now display interactive cards, tables, and other UI components. 

 

Key Advantages:

  1. Build Once, Use Anywhere: Create one widget and reuse it across supported platforms like Agentforce, ChatGPT, Claude, and Slack.
  2. Interactive Responses: Display customer details, records, and other information in a user-friendly format.
  3. Less Development Work: No need to rebuild the same UI for every supported platform.
  4. Custom Lightning Types: Connect Agentforce action outputs to interactive widgets.
  5. Agent Script Integration: Configure action outputs to display widgets in Agentforce.
  6. Security: Use Salesforce permissions and sharing rules to control data access.

Example:

Instead of Agentforce displaying customer details as plain text, we can show the information in a structured card. 

 

My View:

HXL could be very useful for developers building Agentforce solutions across multiple platforms. It's currently in beta, so testing and platform compatibility are important. 

 

Question:

Do you think HXL will reduce the need for building separate custom UI components for AI-powered Salesforce applications? 

 

Useful Resources:

#Agentforce #Salesforce Developer #TrailblazerCommunity #Trailhead Challenges #Salesforce

0/9000

Trailhead: Refine Your Prompt Template for Accurate Agent Replies. 

My agent is working in preview: 

 

Challenge not yet complete in Service Prompt Builder Whoops, looks like there was a problem. Please try again.

 

But still getting past the challenge. Any suggestion? 

 

#Trailhead Challenges  #Agentforce

3 件の回答
  1. 9月25日 6:34

    Hi Tuan, 

     

    I faced the same issue in the Trailhead challenge “Refine Your Prompt Template for Accurate Agent Replies.” 

     

    My Agentforce preview was working correctly, but Trailhead was still showing:

    “Challenge not yet complete. Whoops, looks like there was a problem. Please try again.” 

     

    I fixed it by checking the following steps: 

     

    1. Go to Setup → Flows.
    2. Search for Get Open Cases for Contact.
    3. Open the flow and make sure it is Active.
    4. Go to Setup → Prompt Builder.
    5. Open the prompt template Guest Experience Compensation.
    6. Make sure you are working on the latest version of the prompt template.
    7. In the prompt, after Case Details:, insert: Insert Resource → Flow → Get Open Cases for Contact
    8. After Complaint:, insert: Insert Resource → Free Text → Complaint
    9. Do not manually type the resource names. Insert them using the Insert Resource option.
    10. Save the prompt template.
    11. Test it in Preview with Sofia Rodriguez.
    12. Confirm that the resolved prompt contains:
      • Contact details
      • Open case details
      • Complaint
      • Compensation instructions
    13. Click Activate for the latest version of the Guest Experience Compensation prompt template.
    14. Go back to the Coral Cloud Experience Agent.
    15. Refresh the agent preview and test again.
    16. Use: The White Water Rafting experience my family and I booked was unexpectedly cancelled.
    17. When the agent asks for customer details, enter:  I'm sofiarodriguez@example.com and my membership number is 10008155.
    18. If the agent response is working correctly, go back to Trailhead and click Check Challenge again.

     

    The main things to verify are: 

     

    Get Open Cases for Contact Flow → Active

    Guest Experience Compensation → Latest Version Active

    Case Details → Flow resource inserted correctly

    Complaint → Free Text resource inserted correctly

     

    Thank You, 

    Venkat Yadav.

0/9000

Hi all, I have the following issue in the step 3: 

Einstein is ON, I was able to complete previous steps, but in the step 3, error message all the time: 

 

Unfortunately, Live Chat support is currently unavailable for this agent. Please feel free to reach out to our customer service directly through email or phone, and we'll be happy to assist you.

 

 https://trailhead.salesforce.com/content/learn/modules/setup-with-agentforce-for-admins/manage-user-access-1?trail_id=build-agentforce-solutions-with-low-code-tools

 

#Trailhead Challenges

4 件の回答
  1. 10月9日 17:32

    @Luis Daniel Castañeda Vargas That line isn't a Live Chat outage. It's the fallback when the conversation never reaches an activated agent. Einstein being on doesn't activate the agent.

    In that unit the chat is hitting the agent the step is testing. If the agent is still a draft, or you opened a different one, you get exactly this message. The same thing happens when the user in the playground doesn't have the Agentforce access the earlier steps assigned. The panel opens, and there's no agent behind it.

    Open the agent from this playground, confirm it's the one the step names, and activate it. Then start a new conversation. The chat you already have is a failed session, and refreshing it keeps the same failure.

0/9000

Hi Everyone, 

I’ve created and activated a custom agent called

Campfire Stories in Agentforce Studio. It shows as active, but when I open the Campfire Stories tab in org, the Data 360 Marketing Agent appears instead.

 I checked Agentforce Registry, but both Registered APIs and Registered MCP Servers are empty. Also, Agentforce agent is not showing any dropdown to pick the Campfire Stories agent — it directly gives the Data 360 agent.  I’m unsure how to correctly register or connect my Campfire Stories agent so that it loads properly on the page.

Could anyone guide me on the correct steps to register and connect my custom agent in Agentforce Registry so it opens instead of Data 360?

Thanks in advance! 

 

Campfire Stories Agent not opening, Data 360 Agent shows instead

 

 

Screenshot (672).png

 

 

 

Screenshot (670).png

 

 

 

#Trailhead Challenges

3 件の回答
  1. 10月10日 11:34

    Hi @Hanisha A From your screenshots, the Campfire Stories agent is active, but the chat panel is still showing the Data 360 Marketing Agent. 

     

    Try these steps:

    1. Check Agent Access: Make sure your user has permission to access the Campfire Stories agent.
    2. Check Agent Settings: Verify that the agent is configured for the Employee Agent experience.
    3. Agentforce Registry: You don't need to register the agent under APIs or MCP Servers for this.
    4. Check Agent Assignment: Verify which agent is configured to appear in the Salesforce chat panel.
    5. Refresh and Test: Log out, log back in, and check again.

    Important: Activating a custom agent doesn't automatically make it appear in the Salesforce chat panel.

    Also, make sure you're using the correct Trailhead playground.

    Hope this helps!

0/9000

Hi Trailblazers, 

 

I'm stuck on unit 2 of the Agentforce Configuration for Slack Deployment badge and the checker keeps failing even though I verified every documented step. 

 

Module:

https://trailhead.salesforce.com/content/learn/modules/agentforce-configuration-for-slack-deployment

 

Unit 2:

https://trailhead.salesforce.com/content/learn/modules/agentforce-configuration-for-slack-deployment/configure-a-slack-agent-in-salesforce

 

 

Error returned by "Check Challenge to Earn 500 Points": 

Challenge not yet complete in my hands-on org. 

"We see you're using the Agentforce org from the quick start. But it's not prepared for agent deployment into Slack. Complete the steps from the quick start to ensure you can set up the agent." 

On other attempts it returns simply: "Whoops, looks like there was a problem. Please try again." 

 

What I have completed and verified in the org: 

1. Special Developer Edition org created from the official promo (agentforce-for-slack-on-trailhead) and connected to the hands-on challenge - no org mismatch. 

2. Slack OAuth enabled: Setup > Connected Apps OAuth Usage > Slack > Load Actions > Install (Slack v12.0, App Status Allowed). 

3. All four Slack actions added to the Product Gaps subagent of Product Management Agent: Create a Slack Canvas, Update a Slack Canvas, Search Slack, Send a Slack Direct Message. 

4. Both required instructions added verbatim. 

5. Agent tested in Preview / Live Test - a Product Gap Case was created successfully. 

6. Agent connected to Slack, then Commit Version and Activate (Version 1 Active). 

7. Permission set Access Agents > Agent Access > Product Management Agent enabled and assigned to my user. 

8. Slack Playground reconnected to this org only, my user added and Authenticated, agent installed in the workspace. 

9. My Domain was never changed - it is the default one. 

10. Quick Start: Connect Your Agentforce Org with Slack shows 100% completed (both units). 

 

I already retried with two completely new Developer Edition orgs and a fresh Slack Playground connection. The checker still reports the quick-start org as "not prepared". 

 

Does anyone have a fix, or has anyone else hit this? Any insight on what the checker actually validates would be very helpful. 

 

Thanks! 

 

#Trailhead Challenges

3 件の回答
  1. 10月6日 17:19

    SOLVED - and the culprit was neither the connector package nor caching. Posting this in case it helps the next person. 

     

    The real cause was two toggles in the Quick Start that I had skipped: 

     

    Setup > Quick Find > Einstein Bots  ->  was OFF, I turned it On 

    Setup > Salesforce Go > Search features > Agentforce (Default)  ->  was OFF (still showing Get Started), I did Get Started > Turn On > Confirm 

     

    The Quick Start lists THREE toggles in its "Activate Einstein, Einstein Bots, and Agentforce" section - Einstein Setup, Einstein Bots, and Agentforce (Default). I had only done Einstein and Agentforce Studio. Once both were On, the checker message changed immediately from "...not prepared for agent deployment into Slack" to just the generic "Whoops" message, and after that unit 2 passed. 

     

    Important note: a brand-new org from the agentforce-for-slack-on-trailhead promo does NOT come with Agentforce (Default) already enabled. You have to run that toggle yourself. Einstein Bots is easy to overlook because nothing else in the module mentions it. 

     

    For anyone stuck on unit 3 (Deploy an Agent in Slack), the checker message is very specific: "We can't find the conversation with the Product Management Agent agent. Make sure the messages show up on the Slackbot direct messages." 

     

    Gotcha: a chat with Slackbot does NOT count. You must open the Agentforce Hub / Tools > Agents, open Product Management Agent itself, and start the conversation there. Also give the agent concrete details, otherwise it just replies with clarifying questions and never logs a gap or creates a canvas. A prompt that worked: "Log a product gap: our current product lineup has no analytics or reporting coverage for the new business unit. Contact is Sofia Rodriguez,

    sofiarodriguez@example.com

    . Please create a canvas called Product Gaps with these details." 

     

    Thanks again Vishal - your reply sent me to Installed Packages, and chasing that is what made me re-read the Quick Start and notice the two toggles I had skipped.

0/9000

Hello - I have tried for weeks to complete this challenge to no avail. I am asking for it to be overridden. I have cancelled, deleted, restarted, relaunched, cancelled, over and over and it does not accept my answer to complete this trailblazer. 

 

#Trailhead Challenges

5 件の回答
  1. 10月3日 7:36

    Hi @Charlie Parfet

    I can understand how frustrating this is, especially after trying the same challenge multiple times.

    From your screenshot, the Coral Cloud org and Guest Agent look okay. I would check a few things before trying again:

    • Open reservation R-00000017 and confirm its current status.
    • Make sure the Cancel Reservation action or Flow is active.
    • Confirm that action is added to the correct subagent.
    • Test the cancellation in Agentforce Builder and make sure the reservation record is actually updated.
    • Avoid deleting and recreating the reservation unless the challenge specifically asks you to.
    • Also make sure Trailhead is checking the correct special Developer Edition org.

    If everything is working correctly but the challenge still does not pass, then it could be a Trailhead validation issue. 

     

    Since you have already been trying for weeks, I would suggest raising a Trailhead Help case with the screenshot, challenge name, reservation number, and the exact error message.

    Hope this helps. 

     

    Thank you, 

    Venkat Yadav.

0/9000