Search for accounts/companies in Salesforce
AI agents call search_accounts to retrieve information from Mcp Salesforce without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a search tool that retrieves data from Salesforce without modifying, deleting, or executing any operations. It falls under the Read category as it queries existing accounts. The severity is low because the blast radius of misuse is minimal—an agent could retrieve unwanted account information but cannot alter data or trigger external actions.
From the tool's definition Tool name 'search_accounts' and description 'Search for accounts/companies in Salesforce' indicate a query/retrieval operation with no side effects.
Attacks that exploit this kind of access
Search for accounts/companies in Salesforce. It is categorised as a Read tool in the Mcp Salesforce MCP Server, which means it retrieves data without modifying state.
Register the Mcp Salesforce MCP server in PolicyLayer and add a rule for search_accounts: 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 Mcp Salesforce. Nothing to install.
search_accounts is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the search_accounts 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 search_accounts. 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.
search_accounts is provided by the Mcp Salesforce MCP server (@snokam/mcp-salesforce). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →