Productboard has named the problem that every product organization using AI has now and few have described. In MCP for product teams, the post observes that “every PM on your team now has an AI tool, and each one is using it differently”: specs in Claude, screens in Lovable, code in Cursor, each fed different context. “Two PMs can ask about the same customer and get different answers because their tools have access to different evidence.” Specs change in a coding tool and the roadmap never hears about it. The answer is one sentence: “MCP is how the tools connect. Productboard is the record.”
There are no figures, and the access model is stated carefully. AI tools that support remote MCP can connect to Productboard’s server and “read and edit specs, comment on them, and update their status,” and create and update the product hierarchy from objectives down to subfeatures, along with companies and users. “Each tool signs in with the person’s own Productboard login, so it can only see and do what that person can, and admins decide which roles can use MCP.” The server is included on every plan at no extra cost. The post also describes the other direction: Spark, Productboard’s own agent, connects out to Amplitude, Linear, Notion, Slack and Lovable through MCP, and in some cases acts inside them. The leadership argument is the strongest part: when an executive asks why a feature was prioritized, the answer should not be “spread across private AI chats, specs, tickets, and code comments.”
The catalog holds the record the post describes. The Productboard provider page lists 11 API pages, including the Productboard Entities API and Productboard Entity Relationships API for the product hierarchy, and the Productboard Notes API where customer feedback lands. On the Agent Readiness score, the MCP server and the full identity layer are lit, delegated identity, protected resource metadata and dynamic client registration, which matches the per-user login the post describes.
The Kin Score is 34.9, thin band, with discoverability at 75.0 and contract quality at 46.3 against developer ergonomics at 22.6 and contract governance at 0.0. The Agent Readiness score is 40.3, agent-ready, above the composite. Agentic access is unlit, so the catalog has not mapped which of those operations act, and so are error semantics and reversibility. A system of record that many tools write back to is the place where a bad write is most expensive and an undo is most valuable. Productboard has made itself the shared record for every PM’s AI tool. The contract should say what happens when two of those tools disagree.