FactSet extends MCP suite to portfolio analytics
Buy-side teams get pre-calculated performance, attribution and risk outputs piped into private LLM environments under a limited release.
FactSet has extended its Model Context Protocol (MCP) suite to portfolio analytics, a limited release that routes pre-calculated performance, attribution, and risk outputs into clients' private LLM environments with audit-ready logging, according to the company.
The move pushes FactSet further into the agent-tooling layer that vendors across financial data have been racing to claim since Anthropic published the MCP spec in late 2024. Rather than expose raw datasets to a model and hope the math holds, FactSet is shipping the already-computed analytics, the same numbers a PM or risk officer would pull from a workstation, into whatever LLM the client is running internally.
The audit-ready framing is the part to focus on. Buy-side compliance teams have spent the past 18 months pushing back on "ask the LLM" workflows precisely because the chain of custody on a number, who calculated it, when, from what inputs, breaks the moment a model paraphrases the output. Piping pre-calculated FactSet analytics through MCP keeps the number deterministic and the lineage intact. The LLM becomes the interface, not the calculator.
FactSet has not disclosed pricing, client count for the limited release, or a timeline for general availability.
Are your experts using AI to cheat?
Try our free demo to find out today.
The limited-release structure suggests FactSet is using the launch window to gather workflow feedback before broader rollout. Watch for client disclosures on the Q2 FY2026 earnings call, and for parallel MCP announcements from Bloomberg and S&P Global Market Intelligence in the same window.
Powering institutional-grade transcription for expert networks.
INFLXD provides AI-powered, human-edited transcription with sub-1% error rates for the world's leading expert networks and financial research firms.
Visit inflxd.com →Keep reading.
How Expert Networks Verify Expert Credentials: 7 Structural Models
The pre-panel verification step that determines cost, panel size, and litigation exposure at every major expert network.
Datasite integrates Rogo AI into M&A data rooms
The tie-up lets vetted deal teams run Rogo queries on live Datasite content without exporting files or bypassing room permissions.
Rogo ARR passes USD 50M, tripling year-over-year
The Lazard-alum-founded workflow AI vendor is now running ahead of Hebbia on disclosed revenue after an April Series D marked it at roughly USD 2B.