Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →toddkasper/salesforce-agentforce-specialist
Building and governing Salesforce Agentforce and generative-AI features — Agentforce agents (topics, actions, agent-user security, the reasoning loop), Prompt Builder templates (Sales Email, Field Generation, Record Summary, Flex), Data 360 (formerly Data Cloud) and Knowledge grounding/RAG, and the Einstein Trust Layer (data masking, zero-retention, audit).
Passed 22 of 22 test cases with the skill, and 18 of 22 without it.
and 3 more in the full report
4 fixed18 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
We need to query unstructured content across three separate knowledge repositories and merge the re-ranked results into a single context for prompt grounding. Which retriever type should we select in Data 360?
| Checked | Without skill | With skill |
|---|---|---|
| Recommends selecting an ensemble retriever for multiple unstructured data sources. | Missed | Passed |
The instructions your model reads when the skill loads.
Rebrand note (read first): The original "Salesforce Certified AI Specialist" exam (launched September 2024) was rebranded "Salesforce Certified Agentforce Specialist" (AI-201) on March 3, 2025. Topic areas and weights shifted significantly. Existing AI Specialist credential holders were automatically transitioned with no retest required. "AI Specialist" is no longer the current name — use Agentforce Specialist (AI-201). (Distinct from the entry-level AI Associate credential, which was retired in early 2026.)
The Agentforce Specialist (formerly AI Specialist) validates the ability to implement, configure, and govern Salesforce generative + agentic AI — Prompt Builder, Agentforce agents (formerly Einstein Copilot), Model Builder, Data 360 (formerly Data Cloud) grounding, and the Einstein Trust Layer. This document is an operational playbook, not an exam outline: every section states the rule as an actionable instruction, the real limits, the decision criteria, and the anti-patterns to catch in review.
Product rename (Oct 2025): Salesforce Data Cloud was renamed Data 360 at Dreamforce on October 13, 2025. The underlying product, licenses, data model, and integrations are unchanged. This document uses "Data 360 (formerly Data Cloud)" on first use and "Data 360" thereafter.
The certification sits between entry-level AI Associate (conceptual, no hands-on) and senior architecture credentials. Recommended (not required) prior knowledge: Salesforce Administrator or Platform App Builder.
Load this skill when… building or reviewing Agentforce agents (topics, actions, agent-user security); creating Prompt Builder templates (Sales Email, Field Generation, Record Summary, Flex); configuring Data 360 (formerly Data Cloud) grounding or Knowledge RAG for a prompt or agent; or reviewing Einstein Trust Layer settings (data masking, zero-retention, audit trail). Not this skill: admin-level Agentforce permission setup (enabling features, assigning licenses) without building an agent → see
salesforce-administrator; writing the Apex code behind an agent action → seesalesforce-platform-developer-2.
Deeper context: Study resources live in references/study-resources.md (loaded on demand). For org-specific applications of these rules, see a per-org appendix you maintain in your own project, referenced from a CLAUDE.md. For NPSP/nonprofit-specific guidance, see salesforce-nonprofit-cloud-consultant.
Verify steps assume nothing about your tooling — use your project's Salesforce MCP connection, the Salesforce CLI (
sf), or the Salesforce setup UI, in that order of preference.
Credential logistics and study path: see references/study-resources.md.
test cases passed, vs 18 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.