forcedotcom/sf-skills / plugins/builder/service-engagement/skills/service-agentforce-human-escalation-configure
service-agentforce-human-escalation-configure
UnverifiedUse to configure and verify Agentforce agent-to-human escalation, including human handoff, a staffed fallback queue, and failure-threshold directives. Triggers: configure agent escalation, escalate to a human agent, agent handoff to a queue, configure a human escalation queue, set an Agentforce fallback handoff. Do not use to create an agent or configure inbound channel routing; use service-agentforce-channel-configure for inbound routing.
assets/escalation-thresholds.instructions.md
View on GitHub<!--
Agent Script failure-threshold directive snippet (NGA / Agent Script authoring model).
This is a TEMPLATE of directive text, NOT an enforced runtime counter. It is injected
into the agent's topic/system instructions so the planner escalates consistently.
Salesforce Agentforce does not expose a metadata field that counts failed turns; the
threshold is expressed as instruction text and verified by an ADK eval rubric plus a
documented manual conversation test — never by a deterministic metadata read.
Substitute the tokens below when authoring:
__DEFAULT_FAILURES__ default consecutive-failure threshold (recommended: 2)
__OVERRIDE_TOPIC_LABEL__ label of a topic that needs a tighter threshold (e.g. "Password Reset")
__ESCALATION_MESSAGE__ customer-facing handoff line
-->
# Escalation policy (agent directives)
Add the following to the agent's global/system instructions:
- When you cannot resolve the user's request after __DEFAULT_FAILURES__ consecutive
failed attempts on the same request, escalate to a human agent instead of continuing
to retry.
- Treat "failed attempt" as: you asked a clarifying question and still could not act, a
tool/action returned an error, or the user states your answer did not resolve the
issue.
- On escalation, tell the user: "__ESCALATION_MESSAGE__" and then invoke the human
handoff (the escalation topic is enabled to transfer to the human queue).
- Do not disclose internal error details, stack traces, or system identifiers when
escalating.
# Per-topic override
Add the following to the __OVERRIDE_TOPIC_LABEL__ topic instructions (tighter threshold):
- For __OVERRIDE_TOPIC_LABEL__ requests, escalate to a human after a SINGLE failed attempt.
Time-sensitive requests (for example password/credential problems) should not be
retried repeatedly.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |