The Brev MCP server is a Model Context Protocol server that gives AI agents — like Claude, Cursor, and ChatGPT — live access to your company's goals, OKRs, and operating context. Instead of every prompt starting from a blank slate, agents connected to Brev MCP know what your company is trying to accomplish, who owns what, and how progress is tracked. That goal-awareness gets injected into every response, so AI output stays tied to what actually matters to the business.

Brev MCP exposes your company's goal graph — objectives, key results, owners, cadences, and recent progress — as structured context any MCP-compatible agent can read at runtime. When you prompt an AI tool connected to Brev, it pulls the relevant goal context before responding. The result: agents reason against your real strategy instead of generic assumptions. A prompt like "draft a risk assessment for our next sprint" returns output grounded in your actual Q-level priorities, not boilerplate.

You can read, update, and act on company goals directly from the command line — no UI context-switch required. Typical workflows include checking goal status, posting check-ins, querying owners and cadences, and triggering goal-based automations. The CLI is goal-native: every command is scoped to the goal graph, so AI agents and engineers operate on the same source of truth.

Brev MCP works with any MCP-compatible client, including Claude Desktop, Claude Code, Cursor, ChatGPT desktop, and custom agents built on the OpenAI Agents SDK or Anthropic SDK. Because MCP is a vendor-neutral open standard, support extends to any tool that adopts it — no per-tool integration work required on your end.

System prompts are static; Brev MCP is live. A system prompt captures goals at one point in time and goes stale within days. Brev MCP queries the current state of your goals at the moment of the prompt — including the latest check-ins, status changes, and owner updates. Agents stay aligned with reality instead of a snapshot.

Most teams are connected and pulling goal context into their AI tools in under 10 minutes. Setup is a standard MCP connection: install the server, authenticate once, and your AI client picks up the available tools automatically. No code changes, no custom integrations.

Brev MCP is built for engineering, operations, and go-to-market teams that already use AI tools daily and want those tools to reflect company strategy without manual context-pasting. Common users include founders running lean teams, eng leaders standardizing AI workflows across the org, and operators who want AI output that respects OKRs by default.