When Agentforce performs its grounding check, it examines source information, the topic instructions, and scope. Which additional information does the agent look for during its grounding check?

The correct answer is B. Prompt injection risks. During its grounding check, Agentforce evaluates whether the content it is processing contains prompt injection risks - attempts by malicious text embedded in source data to hijack the agent's instructions or override its intended behavior. This is a critical security step…

Submitted by femi9· Apr 18, 2026Workflow/Process Automation

Question

When Agentforce performs its grounding check, it examines source information, the topic instructions, and scope. Which additional information does the agent look for during its grounding check?

Options

AIncompatible data types
BPrompt injection risks
CEncrypted fields
DThe web for information verification

How the community answered

(23 responses)

B
91% (21)
C
4% (1)
D
4% (1)

Explanation

During its grounding check, Agentforce evaluates whether the content it is processing contains prompt injection risks - attempts by malicious text embedded in source data to hijack the agent's instructions or override its intended behavior. This is a critical security step alongside checking source information, topic instructions, and scope. Incompatible data types, encrypted fields, and web verification are not part of the standard Agentforce grounding check process.

Topics

#AI Security#Prompt Injection#Agent Grounding#Automation Risks

Community Discussion

No community discussion yet for this question.

Full PLAT-ADMN-201 Practice