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/prompt-template-action.agent
View on GitHub# Prompt Template Action Pattern
# Invoke Salesforce PromptTemplates directly from Agent Script
#
# ★ When To Use This Pattern:
# - Generate dynamic AI content (summaries, emails, recommendations)
# - Use Einstein Prompt Builder templates from agents
# - Combine structured data with LLM generation
# - Leverage existing PromptTemplate investments
#
# ★ Key Insights:
# - Target uses generatePromptResponse:// protocol (not flow://)
# - Input names MUST have "Input:" prefix (critical!)
# - Output field is always named "promptResponse"
# - PromptTemplate must be deployed BEFORE the agent
#
# ★ Important Limitation:
# - Template API name must exactly match target
# - Record-bound templates require recordId input
#
# This is a COMPLETE template - customize for your use case
system:
instructions: "You are a content generation assistant. Help users create personalized content using AI templates. Always verify the generated content meets quality standards."
messages:
welcome: "Hello! I can help generate personalized content for you."
error: "I encountered an issue generating content. Let me try again."
access:
default_agent_user: "agent@company.salesforce.com"
config:
developer_name: "Content_Generator_Agent"
agent_label: "Content Generator"
description: "Agent that generates AI content using Prompt Templates"
variables:
# Standard linked variables
EndUserId: linked string
source: @MessagingSession.MessagingEndUserId
description: "Messaging End User ID"
RoutableId: linked string
source: @MessagingSession.Id
description: "Messaging Session ID"
ContactId: linked string
source: @MessagingEndUser.ContactId
description: "Contact ID"
language:
default_locale: "en_US"
start_agent agent_router:
label: "Content Generator"
description: "Routes to content generation options"
reasoning:
instructions: ->
| Ask the user what type of content they'd like to generate:
1. Personalized daily schedule
2. Professional email
Route to the appropriate request handler.
actions:
go_schedule: @utils.transition to @subagent.schedule_generation
go_email: @utils.transition to @subagent.email_generation
subagent schedule_generation:
label: "Schedule Generation"
description: "Generate personalized daily schedules"
actions:
# ★ PROMPT TEMPLATE ACTION - Key syntax elements:
# 1. target: uses generatePromptResponse:// protocol
# 2. inputs: MUST use "Input:" prefix
# 3. outputs: field name is always "promptResponse"
Generate_Personalized_Schedule:
description: "Generate a personalized daily schedule using AI."
inputs:
# ★ CRITICAL: Input names must have "Input:" prefix
# These map to {!email} and {!timezone} in the PromptTemplate
"Input:email": string
description: "User's email address for preference lookup"
is_required: True
"Input:timezone": string
description: "User's timezone for schedule formatting"
is_required: False
outputs:
# ★ Standard output field name for prompt responses
promptResponse: string
description: "The AI-generated schedule content"
is_used_by_planner: True
is_displayable: True
# ★ Target protocol: generatePromptResponse://TemplateApiName
target: "generatePromptResponse://Generate_Personalized_Schedule"
reasoning:
instructions: ->
| Help the user create a personalized daily schedule.
1. Ask for their email (required)
2. Ask for their timezone (optional, defaults to Pacific)
3. Generate the schedule using the AI template
4. Present the generated schedule
The schedule will include morning routine, work blocks,
breaks, and evening wind-down based on preferences.
actions:
create_schedule: @actions.Generate_Personalized_Schedule
# ★ Input binding with quoted field names
with "Input:email"=...
with "Input:timezone"=...
back: @utils.transition to @subagent.agent_router
subagent email_generation:
label: "Email Generation"
description: "Generate professional emails"
actions:
# Another Prompt Template action example
Generate_Professional_Email:
description: "Generate a professional email draft."
inputs:
"Input:recipientName": string
description: "Name of the email recipient"
is_required: True
"Input:topic": string
description: "Topic or purpose of the email"
is_required: True
"Input:tone": string
description: "Desired tone: formal, friendly, urgent"
is_required: False
outputs:
promptResponse: string
description: "The generated email content"
is_used_by_planner: True
is_displayable: True
target: "generatePromptResponse://Generate_Professional_Email"
reasoning:
instructions: ->
| Help the user draft a professional email.
Collect:
1. Recipient's name (required)
2. Email topic/purpose (required)
3. Desired tone: formal, friendly, or urgent (optional)
Generate the email and present it for review.
actions:
create_email: @actions.Generate_Professional_Email
with "Input:recipientName"=...
with "Input:topic"=...
with "Input:tone"=...
back: @utils.transition to @subagent.agent_router
# ★ Insight: Prompt Template Integration Checklist
#
# BEFORE PUBLISHING AGENT:
# 1. Create PromptTemplate metadata file
# 2. Deploy PromptTemplate: sf project deploy start -m "PromptTemplate:Template_Name"
# 3. Verify template exists: sf org list metadata -m PromptTemplate
#
# INPUT MAPPING:
# Agent Script "Input:fieldName" → PromptTemplate {!fieldName}
# "Input:email" → {!email}
# "Input:customerName" → {!customerName}
#
# COMMON ERRORS:
# - HTTP 404: PromptTemplate not deployed yet
# - "property X not found": Input name doesn't match template variable
# - Missing "Input:" prefix: Inputs won't map correctly
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 |