Agentblazer Community Group

"Give them System Admin for now." 🙃 

Five of the most expensive words in Salesforce. Because "for now" never seems to end, and nobody ever raises a ticket asking to have their access removed.  

This Sunday in Session 2, Trust, Earned Not Assumed, we'll roll up our sleeves in a live org: 

🗝️ A "who has the keys?" audit: admins, integration users, sharing defaults, guest users  

🛡️ How to scope an AI agent so it only touches what it should  

💬 Why the strongest defence against prompt injection is surprisingly simple

You don't need to be a security expert. You just need to care about the people whose data lives in your org. 

And trust me, you'll leave wanting to check your org first thing Monday morning. 😄

🗓️ Sunday 18 Oct · 11:00 GST · Online  

👉 Register : https://trailblazercommunitygroups.com/j/ntzrw63pjpn94/

@Salesforce Developer Group, Dubai, UAE @Agentblazer Community Group @Architect Trailblazers @* Salesforce Developers *

 👉 Click here to RSVP

For years, my answer to is this a good architecture? was three confident words: scalable, secure, maintainable.  

Nobody ever asked me what those words actually meant. Including me. This Sunday, that changes. We're kicking off a 7-week live series on the new Salesforce Well-Architected Framework, and Session 1 is all about the big picture: 

• ✨ What changed, and why 3 pillars became 5 
• ✨ Why this is NOT just the AI framework 
• ✨ A live 5-question health check you can run on your own org on Monday morning

You'll leave with a clear map of what good really looks like on Salesforce, and a language your whole team can share. Come for the pentagon. Stay for the aha moments. 📐 

@Salesforce Developer Group, Dubai, UAE @Agentblazer Community Group @Architect Trailblazers @* Salesforce Developers * 

🚀 Welcome to #DF26LearningQuest🚀 

A massive THANK YOU to everyone who has already joined the quest! 💙 

To clear up any quick questions: This is NOT a coding competition. There are no speed tests or code challenges here. Our goal is simple: to come together as a Trailblazer community, attend #Dreamforce (in person or virtually), and maximize our learning journey side by side. 🧠✨ 

How to Submit Your Posts:

1. Create your social media post using the hashtag: #DF26LearningQuest.
2. Submit your post link using the form: https://forms.gle/VmDu2EsH53DgEpJ16
3. ⚠️ Important: Please use the same email address you used when registering for the quest so we can log your entries accurately.

Here is your official roadmap to participate and share your experience 👇

How to Participate: 

1️⃣ Start Posting Now

• Announce that you’re participating in #DF26LearningQuestand invite your network to join you! 

2️⃣ Share How to Register for Free

• Help fellow Trailblazers navigate free registration options and streaming via Salesforce+. 

3️⃣ Guide Agenda Building

• Share your must-attend session picks, keynotes, and hands-on labs to help others build their schedules. 

4️⃣ Share Live Updates During Dreamforce

• Post about the sessions you attend during the event, tagging speakers and key topics. 

5️⃣ Post Key Takeaways & Aha! Moments

• Highlight the biggest announcements, AI updates, and technical insights you take away from each session. 

6️⃣ Keep the Momentum Going Post-Dreamforce

• Continue sharing recap threads, Trailhead trailmix recommendations, and session highlights after the main event.

7️⃣ Attend the 2 Post-Dreamforce Live Sessions

• Join the 2 live post-event sessions to deep-dive into major announcements, then share your reflections!

⚠️  Official Disclaimer & Program Terms :

• No Guarantee of Reward: Participation in #DF26LearningQuest does not guarantee the receipt of a Salesforce certification voucher or any other reward. Voucher quantities are strictly limited.
• Review & Selection Criteria: All submitted social media posts will be manually reviewed by the organizing committee. Submissions will be evaluated based on originality, quality of content, community engagement, and practical learning takeaways. Low-effort posts, spam, duplicated content, or posts failing to demonstrate genuine engagement will be disqualified.
• Form & Email Verification: To be eligible for review, participants must post publicly using the #DF26LearningQuesthashtag and submit the post link via the official submission form using the exact email address used during initial quest registration. Unsubmitted links or posts submitted with mismatched emails will not be tracked.
• Discretion & Finality of Decisions: Voucher allocation is strictly at the sole discretion of the organizing committee. All committee rulings regarding submission validity, quality scoring, and voucher distribution are final, binding, and non-negotiable.
• Voucher Terms & Non-Transferability: Certification vouchers (if awarded) are non-transferable, cannot be resold or exchanged for cash, and are subject to Salesforce's official voucher expiration dates and exam policies.
• Community Initiative Notice: This quest is an independent community-driven educational initiative organized by Trailblazer Community Leaders. It is not directly managed, sponsored, or legally guaranteed by Salesforce, Inc.
• Content Usage: By submitting a link to your post, you grant the organizing committee permission to re-share, quote, or feature your public submission in community recaps and social media highlights. 

#DF26LearningQuest #Dreamforce #Salesforce @Trailblazer Community Cove @Salesforce Developer Group, Dubai, UAE @Agentblazer Community Group @Architect Trailblazers

Two architects walk into a design review. 🏛️ 

• Architect A: "Make it asynchronous." 
• Architect B: "No, synchronous is fine." 

Twenty minutes later, they're still arguing. Neither is wrong. Neither has said what they actually care about. The meeting is now running over, and the winner is whoever is loudest or has the most grey hair. 

That's exactly where our first live session started: "good architecture" used to be a vibe. Then Salesforce rebuilt its Well-Architected Framework. Three pillars became five: Trust, Reliability, Operational Excellence, Resource & Cost Optimization, and Fairness. Now everyone in the room gets the same questions and the same vocabulary, so the argument becomes a conscious trade-off instead of a shouting match. 

A big thank you to everyone who joined, asked sharp questions, and kept the discussion going. You made the session a lot more fun than any design review I've seen. 🙌 

Missed it? No stress. The recording is waiting for you here 👇 

https://youtu.be/Yias41SqkmE 

## Context
Building a POC Agentforce for Service on a production org (Enterprise Edition, Lightning Knowledge, Data Cloud enabled, Foundations active).
**Agent**: Employee Agent (`BotType = AgentforceEmployeeAgent`, `PlannerType = AiCopilot__ReAct`, `BotUserId = null` → runtime user = logged-in user, who is System Administrator with full Knowledge perms).
**Goal**: technician asks "How do I configure the VPN FortiClient?" → agent retrieves the relevant Knowledge article and answers.
## Problem
The action `Answer Questions with Knowledge` (`EmployeeCopilot__AnswerQuestionsWithKnowledge`, `invocationTarget = streamKnowledgeSearch`) **executes successfully but returns 0 results silently**:
```json
Output: {"knowledgeSummary": "Sorry, I can't find an answer based on the available articles."}

The same query via direct SOSL on Knowledge__kav returns the articles correctly — so it's NOT a Knowledge data problem, it's the RAG pipeline.

Event log input observed:

{
  "citationsEnabled": "true",
  "ragFeatureConfigId": "AiRagFeaturePromptContent_e3a977c6_c149_44c0_81fe_a6876f452fd2",
  "query": "Comment configurer le VPN FortiClient ?"
}

The ragFeatureConfigId GUID is the same since the Data Library was first created — even after delete + recreate of the library, Salesforce reuses the same backend record.

What's verified OK

• 4 Knowledge articles, PublishStatus=Online, Language=fr, IsVisibleInApp=true
• Articles are RecordType=Procedure with content in Procedure_Steps__c (10–33k chars each, real procedure steps)
• Data Library assigned to the agent (Knowledge type, default Data Space)
• Content Fields include Procedure_Steps__c and all Procedure_*__c
• Identifying Fields: Title + Summary
• Filter by Knowledge Data Categories: OFF
• Use Public Knowledge Articles: ON

Search Index state (via SOQL on DataSemanticSearch):

• RuntimeStatus = READY
• SearchType = Hybrid (semantic + keyword)
• IndexRefreshedOn recent (post-publish)
• 290 chunks in *_chunk__dlm, 13 contain "FortiClient" literally
• 290 vectors in *_index__dlm
• Max chunk length: 2104 chars (Procedure_Steps content confirmed indexed)
• SharingStatus = REGULAR

Things I tested (none fixed it)

1. Toggle Filter Data Categories OFF (was a forum-suggested cause)
2. Created Data Category Group, tagged all 4 articles, configured Data Category Visibility "All" on the user profile + Default Visibility
3. Topic language changed from en_US to fr (matches articles)
4. Topic instructions translated to French and customized (was the generic English SvcCopilotTmpl__GeneralFAQ template)
5. Activate/Deactivate cycle on the agent
6. Rebuild of the Search Index (SUBMITTED → IN_PROGRESS → READY)
7. Delete + recreate the Data Library → Salesforce reuses the same DataSemanticSearch GUID + same AiRagFeaturePromptContent_* GUID (singleton per data space? bug?)
8. Tried REST endpoints to query the index directly in v60/v62/v66: /ssot/queries/semantic-search, /ssot/queries/hybrid-search, /ssot/retrievers → all 404

Other relevant settings

• Knowledge.settings: enableKbStandardSharing = false (legacy sharing model)
• Knowledge.settings: defaultLanguage = fr (matches articles)
• Profile (System Admin): AllowViewKnowledge = true, EditKnowledge = true, ManageKnowledge = true
• Note: Einstein Agent license users in the org have UserPermissionsKnowledgeUser blocked by license (not used at runtime here since Employee Agent uses logged-in user — but mentioning for completeness)

Questions for the community

1. The AiRagFeaturePromptContent_* GUID — how do I inspect/debug its actual configuration? Is there a Setup UI, a Tooling API object, or a REST endpoint to retrieve its parameters (similarity threshold, top_k, default filters)?
2. Is it expected that delete + recreate of a Data Library reuses the same backend DataSemanticSearch and AiRagFeaturePromptContent? If yes, how does one force a clean reset of the backend RAG config?
3. Does the Data Cloud Hybrid retriever respect the legacy Knowledge sharing model (enableKbStandardSharing=false), the modern one, or its own independent permission layer?
4. What's the correct REST path for direct hybrid/semantic search against a DataSemanticSearch index (outside the agent context, for debugging)? /ssot/queries/semantic-search 404s in API v60-v66.
5. Has anyone else seen this exact symptom — Search Index Ready, content indexed and keyword-matching, agent returns 0 silently — and what was the fix?

Thanks for any pointers!