LogoClawIndex
CasesSkillsAbout
LogoClawIndex

agentforce-guardrails - Define and troubleshoot Agentforce guardrails

Defines, audits, and troubleshoots Agentforce behavioral boundaries using topic scope, system instructions, filters, escalation, and adherence monitoring.

Tags

Updated: 2026-10-07
agentforceguardrailstopic scopeagent instructionsescalationaction filtersrestricted topicsinstruction adherenceabuse preventioncustomer-facing agents

Capabilities

Define topic scope boundariesConfigure agent fallback instructionsApply topic and action filtersConfigure escalation routing

Typical Inputs

Agent name and deployment channelAllowed and refused topicsConfigured topic actions

Typical Outputs

Revised topic Scope textAgent system instruction textFilter configuration decisions

What this skill does

  • Define topic scope boundaries
  • Configure agent fallback instructions
  • Apply topic and action filters
  • Configure escalation routing
  • Review restricted topics
  • Monitor instruction adherence
  • Improve abuse prevention

Inputs

  • Agent name and deployment channel
  • Allowed and refused topics
  • Configured topic actions
  • Human escalation requirements
  • Omni-Channel routing configuration
  • Topic Scope text
  • Classification Description text
  • Instruction Adherence data

Outputs

  • Revised topic Scope text
  • Agent system instruction text
  • Filter configuration decisions
  • Escalation configuration checklist
  • Adherence improvement recommendations
  • Abuse prevention instruction pattern

Requirements

  • Salesforce Spring '25 or later
  • Agentforce org access
  • Agentforce Analytics access for monitoring
  • Omni-Channel configuration for escalation

Source

  • Spec: SKILL.md

ClawIndex

OpenClaw Skills & Use Case Index

ClawIndex is an ecosystem-driven index of OpenClaw skills and real-world use cases.

Index

Skills·
Cases

Meta

About·
Disclaimer·
Email·
GitHub
© 2026 ClawIndex All Rights Reserved.