Guideline schedules U.S. ad-pricing queries in its AI Agent and MCP for October 30
The release’s dated availability table puts AI‑accessible U.S. pricing first, followed by market and buying‑channel data updates in November and December. Access depends on the client’s subscription.
U.S. pricing is scheduled for the AI Agent and MCP on October 30
Guideline has announced U.S. digital advertising pricing data for its AI Agent and Ad Intelligence MCP Server, with availability scheduled for October 30, 2026. Its October 28 release describes two ways for clients to ask about the same data: through Guideline’s own agent, or through an assistant connected to its MCP server.
The intended query is concrete: what digital advertising costs by publisher, ad type or programmatic marketplace, with pricing and spend returned together. Guideline also describes configurable dashboards and charting for dimensions, measures, filters and formulas. Those are announced product functions, rather than independently tested outputs.
The release uses “now” in its descriptive copy, but its explicit availability table dates this AI Agent/MCP enhancement to October 30. As of this October 29 report, that dated schedule is the useful planning reference; the copy does not establish access for every client today.
The rollout has separate dates for separate datasets
Buying channel and geography answer different planning questions
The later Programmatic Direct view is intended to separate U.S. spend and pricing into Direct, Programmatic Direct and Third‑Party buying channels. Guideline explains that a buy placed directly with a publisher can still run programmatically, including private‑marketplace or programmatic‑guaranteed deals. The new view is therefore about how inventory was bought, rather than simply which country supplied the data.
The Canadian retail‑media update instead isolates that market’s retail‑media spend. The U.K. and Canadian classification update aligns digital subtype names with the U.S. scheme across current and historical data. For a team comparing reports, the practical questions are which market, buying channel and classification version a result uses. A shared query interface does not make those dimensions interchangeable.
An assistant connection is a route to entitled data
Related News
Oct 29, 2026
Fibr’s AXO separates website information for AI visitors from agent actions
Fibr says its new Agent Experience Optimization adapts existing approved website information for AI visitors and supports agent-led tasks. Content delivery and transaction permissions remain different questions for buyers.
Oct 29, 2026
Bivia’s typed decision API separates an answer from an abstention
Articul8’s new decision model returns structured values for agent workflows. Its response contract distinguishes answered results from null-valued abstentions, and developers can disable abstention.
Oct 29, 2026
Microsoft’s Copilot harness comes to Foundry for managed agent execution
Microsoft’s October 28 announcement separates the model from its instructions, context, execution and controls, with managed Foundry and downloadable SDK routes.