apply_microsoft_ads_recommendations

Apply Microsoft Advertising recommendations by id. THIS SPENDS REAL MONEY AND HAS NO UNDO. It raises daily budgets, adds keywords, widens keywords to broad match, DELETES negative keywords so the account buys more searches, and publishes Microsoft-written ads under the brand’s name, all live at the next auction. THE GATE IS UNLIKE THE OTHER MICROSOFT SPEND SWITCHES, AND confirm:true ALONE IS REFUSED: it proves you meant to apply something, not that you know what THESE ones do, and it reads identically for one keyword suggestion and forty budget raises. So call it ONCE with no confirmation. Nothing is applied, and you get every recommendation named with its campaign, exactly what it changes, the current -> recommended budget where there is one, and Microsoft’s own cost estimate. SHOW THAT LIST TO THE USER, get an explicit yes, then call again with confirm:true, confirmCount:<n> and confirmCostIncrease:<n>. BOTH NUMBERS ARE RECOMPUTED FROM A FRESH READ at that moment, so a recommendation Microsoft has withdrawn, already applied or re-priced since you looked fails the check instead of being applied unseen. If ANY id you name is no longer on offer, NOTHING is applied, not even the ones that still are, because applying part of a set you inspected whole is not what you asked for. Microsoft can reject individual items and apply the rest, so the result reports per item: never call a partial apply a failure, and never call it a success. 0 credits to us; the spend lands on the ad account. It is categorised as a Write tool in the Hermoso MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

Register the Hermoso MCP server in PolicyLayer and add a rule for apply_microsoft_ads_recommendations: 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 Hermoso. Nothing to install.

apply_microsoft_ads_recommendations 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 apply_microsoft_ads_recommendations 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 apply_microsoft_ads_recommendations. 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.

apply_microsoft_ads_recommendations is provided by the Hermoso MCP server (https://app.hermoso.ai/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.