sf-ai-agentforce for Openclaw

A professional development skill for managing Agentforce Builder metadata, declarative topics, and AI-driven actions within the Salesforce ecosystem.

dsouza-anush
v1.0.0
Apr 9, 2026
0
744
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install sf-ai-agentforce

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install sf-ai-agentforce using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is sf-ai-agentforce?

sf-ai-agentforce is a specialized toolset designed for the Setup UI and Agent Builder path in Salesforce development. It focuses on maintaining Builder-managed topics, actions, and Prompt Builder templates stored as metadata. By integrating these Openclaw Skills into your workflow, you can effectively manage the lifecycle of Service and Employee Agents, ensuring that complex GenAiFunction and GenAiPlugin metadata are correctly configured and deployed.

This skill is particularly valuable for teams transitioning from declarative setup to version-controlled metadata. It provides the necessary bridge between the Agentforce Builder and the technical implementation of Models API, Apex actions, and Flow integrations, making it a cornerstone for modern AI agent orchestration.

sf-ai-agentforce Use Cases

  • Maintaining existing Builder-based agents via Setup and the Agentforce Builder.
  • Creating or troubleshooting GenAiFunction, GenAiPlugin, and GenAiPromptTemplate metadata files.
  • Mapping Builder topics to complex back-end logic such as Flow, Apex, or Prompt Builder actions.
  • Implementing Models API or LightningTypeBundle configurations within an AI-driven Salesforce environment.
  • Managing visibility for Employee Agents via automated permission set configurations.

How sf-ai-agentforce Works

  1. Confirm the project scope specifically targets the Agentforce Builder or Setup UI metadata path.
  2. Determine the agent persona, distinguishing between a Service Agent and an Employee Agent.
  3. Provision the required infrastructure, such as creating a running user for Service Agents or permission sets for Employee Agents.
  4. Define high-quality topics with distinct scope and scenario-based instructions to guide the AI planner.
  5. Assemble supporting actions by wrapping Flows, Apex classes, or Prompt templates into GenAiFunctions.
  6. Validate all metadata dependencies and publish the agent to create a deployable version.
  7. Use command-line tools to activate the specific agent version for end-user access.

sf-ai-agentforce Setup

To get started with these Openclaw Skills, ensure your Salesforce environment is running API v66.0 or higher. For Service Agents, provision the running user using the Salesforce CLI:

sf org create agent-user --target-org <alias> --json

Once metadata is deployed and the agent is published, activate the agent to make it live:

sf agent activate --api-name <AgentName> --version <version_number> --target-org <alias> --json

sf-ai-agentforce Data Schema & Taxonomy

The skill manages standard Salesforce metadata types organized in the following taxonomy:

Metadata Type Directory File Suffix
GenAiFunction genAiFunctions/ .genAiFunction-meta.xml
GenAiPlugin genAiPlugins/ .genAiPlugin-meta.xml
GenAiPromptTemplate genAiPromptTemplates/ .genAiPromptTemplate-meta.xml

Each action is registered as a wrapper around a target (Flow, Apex, or Template), ensuring that input/output contracts match the underlying logic execution.

sf-ai-agentforce Advanced Features

  • Support for Models API integration within Apex-driven AI orchestration.
  • Automated permission set management for Employee Agent visibility using the agentAccesses tag.
  • Advanced topic routing optimization via scenario-based planning descriptions.
  • Integration with custom Lightning Types for structured data input and rich UI presentation.
  • Cross-skill orchestration enabling seamless handoffs between sf-flow, sf-apex, and sf-deploy.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*