forcedotcom/sf-skills / plugins/builder/agentforce-adlc/skills/agentforce-generate
agentforce-generate
FlaggedBuild, modify, audit, repair, optimize, debug, and deploy agents with Agentforce Agent Script. TRIGGER when: user creates, reviews, or changes .agent files or aiAuthoringBundle metadata; asks to fix AgentScript, audit an existing agent, run an AgentScript health check, common-pitfall review, or baseline-versus-candidate repair loop; changes a response, action, subagent, route, state flow, or Agent Spec; migrates or upgrades a legacy GenAiPlannerBundle Agentforce agent (NOT an Einstein Bot) to an aiAuthoringBundle (Agent Script) agent; previews, debugs, deploys, publishes, or tests agents; uses sf agent generate/preview/publish/test; or manages Agentforce MCP servers, tools, assets, or authentication. DO NOT TRIGGER when: Apex, Flow, Prompt Template, Experience Cloud, or general Salesforce CLI work is unrelated to Agent Script; the source is an Einstein Bot (BotDefinition/BotVersion) — use agentforce-bot-upgrade; or the primary input is a production session or trace ID rather than an agent artifact.
assets/patterns/delegation-routing.agent
View on GitHub# Subagent Delegation vs Transition Pattern
# Understanding the difference between delegation and permanent handoffs
#
# ★ KEY CONCEPTS:
#
# DELEGATION (@subagent.* syntax):
# - Syntax: action_name: @subagent.subagent_name
# - Control CAN return to the calling subagent
# - Use for: consulting specialists, sub-tasks, getting help
#
# TRANSITION (@utils.transition syntax):
# - Syntax: action_name: @utils.transition to @subagent.subagent_name
# - PERMANENT handoff - control does NOT return
# - Use for: menu navigation, workflow stages, permanent routing
#
# ★ When To Use Each:
# DELEGATION: "Consult an expert then continue here"
# TRANSITION: "Go to this subagent and stay there"
#
# ★ Implementation Note:
# If delegation syntax does not work in your deployment method, use explicit
# transitions. For a fixed caller, transition directly back without state.
#
# This is a COMPLETE template - customize metadata and task instructions.
system:
instructions: |
You coordinate general and specialist help. Delegate a bounded
consultation when its result is needed in the current task. Use a
transition only when the user's work should continue in another
request handler.
messages:
welcome: "Hello! How can I help?"
error: "I couldn't complete that request."
access:
default_agent_user: "agent@company.salesforce.com"
config:
developer_name: "Delegation_Routing_Agent"
agent_label: "Delegation and Transition"
description: "Demonstrates returning delegation and one-way transitions"
language:
default_locale: "en_US"
start_agent main_hub:
label: "Main Help"
description: "Coordinates general requests and bounded expert consultations"
reasoning:
instructions: ->
| Answer general questions directly. For a bounded expert question,
delegate to the specialist and use the returned answer. For an
order request, transition to order help.
actions:
consult_expert: @subagent.specialist
description: "Consult a specialist, receive the result, and continue the current task"
go_orders: @utils.transition to @subagent.orders
description: "Move permanently to order help"
subagent specialist:
label: "Specialist"
description: "Answers bounded expert questions delegated by the main help task"
reasoning:
instructions: ->
| Provide a focused answer to the delegated expert question. Do not
broaden the task or route elsewhere.
subagent orders:
label: "Order Help"
description: "Handles order questions after a one-way transition"
reasoning:
instructions: ->
| Help with the user's order question. If the user changes intent,
return to general request handling.
actions:
back_to_main: @utils.transition to @subagent.main_hub
description: "Return after the user changes intent"
# ═══════════════════════════════════════════════════════════════
# PATTERN 1: Subagent Delegation (can return)
# ═══════════════════════════════════════════════════════════════
# Delegation syntax in reasoning.actions:
#
# reasoning:
# actions:
# # Delegation - specialist CAN return control to this subagent
# consult_expert: @subagent.specialist_subagent
# description: "Consult specialist for complex questions"
# available when @variables.needs_expert_help == True
#
# The specialist subagent processes the request and control returns
# to the original subagent when done.
# ═══════════════════════════════════════════════════════════════
# PATTERN 2: Transition (permanent handoff)
# ═══════════════════════════════════════════════════════════════
# Transition syntax in reasoning.actions:
#
# reasoning:
# actions:
# # Transition - permanent move to orders subagent
# go_orders: @utils.transition to @subagent.orders
#
# Control moves to orders subagent and STAYS there.
# User continues in that subagent until another transition occurs.
# ═══════════════════════════════════════════════════════════════
# PATTERN 3: Explicit Fixed Return (fallback pattern)
# ═══════════════════════════════════════════════════════════════
# If delegation is unavailable and the caller is fixed:
#
# # In the caller:
# go_specialist: @utils.transition to @subagent.specialist
#
# # In the specialist:
# return_home: @utils.transition to @subagent.main_hub
#
# Store a dynamic return address only when one specialist serves multiple
# callers and deterministic logic reads, clears, and resets that address.
# ═══════════════════════════════════════════════════════════════
# COMPARISON TABLE
# ═══════════════════════════════════════════════════════════════
#
# | Feature | Delegation | Transition |
# |----------------------|-------------------|-------------------------|
# | Syntax | @subagent.name | @utils.transition to |
# | Returns to caller? | YES | NO |
# | Use in actions: | YES | YES |
# | Use in lifecycle: | NO | YES (bare syntax) |
# | Best for | Consult & return | Menu/workflow routing |
#
# ═══════════════════════════════════════════════════════════════
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |