sf_create_agent
Creates an Agentforce Agent (Einstein Copilot) shell in Salesforce. THIS IS STEP 1 OF A 5-STEP SEQUENCE — calling this tool alone produces a non-functional agent shell. You MUST continue calling tools in this exact order without stopping: (0) FIRST create the backing flows (sf_create_flow, flowTy...
This record as markdown: /tools/io-github-semwalajay83-sem-salesforce-metadata-mcp/sf-create-agent.md
What sf_create_agent does on Salesforce Metadata Mcp
AI agents use sf_create_agent to create or update resources in Salesforce Metadata Mcp, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Salesforce Metadata Mcp environment.
Why sf_create_agent is rated Medium
This tool creates new Agentforce Agent configurations in Salesforce, which is a reversible metadata write operation. It does not delete data (not Destructive), does not execute arbitrary code or external operations (not Execute), does not move money (not Financial), and is not a simple read (not Read).
From the tool's definition 'Creates an Agentforce Agent (Einstein Copilot) shell in Salesforce' — the tool explicitly creates a new configuration object that modifies Salesforce's agent setup.
Attacks that exploit this kind of access
The rule that runs sf_create_agent safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Salesforce Metadata Mcp, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For sf_create_agent, this is the rule to start with:
sf_create_agent stays usable, but capped: an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Salesforce Metadata Mcp, apply this rule, and every sf_create_agent call is checked against it from then on.
Questions about sf_create_agent
Creates an Agentforce Agent (Einstein Copilot) shell in Salesforce. THIS IS STEP 1 OF A 5-STEP SEQUENCE — calling this tool alone produces a non-functional agent shell. You MUST continue calling tools in this exact order without stopping: (0) FIRST create the backing flows (sf_create_flow, flowType=AutoLaunchedFlow, status=Active) and/or Apex classes (sf_create_apex_class with @InvocableMethod) for each capability. (1) THIS TOOL — sf_create_agent. (2) sf_create_agent_action — one call per flow/apex. (3) sf_create_agent_topic — group actions into topics; pass all action API names in the. It is categorised as a Write tool in the Salesforce Metadata Mcp MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Salesforce Metadata MCP server in PolicyLayer and add a rule for sf_create_agent: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Salesforce Metadata Mcp. Nothing to install.
sf_create_agent is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the sf_create_agent rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for sf_create_agent. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
sf_create_agent is provided by the Salesforce Metadata MCP server (salesforce-metadata-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Salesforce Metadata, and thousands of servers like it.
This server
Across the catalogue