AI agents call get_share_holdings to retrieve information from Akahu without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves financial data (investment holdings and portfolio value) but does not modify, execute, or move money. It is a Read operation, but severity is high because the data exposed includes sensitive personal financial information about investment holdings and portfolio value, which could be misused if an agent improperly shares or processes this data. No ability to execute transactions or modify accounts.
From the tool's definition Tool returns user's investment portfolio data including total value. Description states 'Return the user's Sharesight investment portfolio: total value' — a retrieval operation with no modification or execution.
Attacks that exploit this kind of access
Return the user's Sharesight investment portfolio: total value,. It is categorised as a Read tool in the Akahu MCP Server, which means it retrieves data without modifying state.
Register the Akahu MCP server in PolicyLayer and add a rule for get_share_holdings: 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 Akahu. Nothing to install.
get_share_holdings 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 get_share_holdings 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 get_share_holdings. 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.
get_share_holdings is provided by the Akahu MCP server (nzkea/akahu-mcp). 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 →