Greenhouse released its Model Context Protocol on May 7, 2026, providing a permission-aware connection layer that enables external AI tools and agents to integrate with the hiring platform while maintaining governance controls. The MCP addresses a specific gap: organizations want to experiment with AI in recruiting without creating compliance risk or abandoning their system of record.
MCP tools operate within existing user permissions. All access flows through audit trails with organization-level controls, rate limits, and safety guardrails. Design partners StubHub and Komodo Health participated in beta testing. At Komodo Health, the protocol accelerated pipeline analytics that previously required dedicated BI teams, making those insights deliverable in under 30 minutes.
The practical applications span automation of QBR summaries, pipeline analysis, and offer digests; complex cross-system investigations blending Greenhouse and HRIS data; and internal hiring copilots operating within Slack or Teams. Rollout to general availability begins June 2026.
Chief Product Officer Meredith Johnson framed the design principle: “AI should strengthen hiring, not shortcut it.” Matt Texeira of Komodo Health added: “The MCP server democratizes access to recruiting intelligence.”
Related: AI Is Quietly Breaking the Signals Hiring Depends On
