MCP (Model Context Protocol)
Why it matters
MCP is the bridge between strategic infrastructure and AI execution — it enables agents to query current intent rather than relying on stale training data or static prompts. By making strategy queryable via protocol, MCP transforms AI agents from context-blind executors to strategically-aligned operators.
How Stratafy addresses this
Stratafy implements MCP as a live bridge between strategic infrastructure and AI agents. The MCP server exposes queryable tools, named resources, and prompt templates — giving any connected AI agent structured, permission-gated access to the full strategic layer in real time.
Queryable strategy tools
The MCP server exposes tools like get_strategy_tree, list_strategies, get_workspace_snapshot, and get_strategy. An agent asks "What are our active strategies?" and gets a structured, real-time answer — including sub-strategies, initiatives, objectives, risks, and assumptions in a single call.
13 prompt templates for strategic co-working
Beyond query tools, the server registers named prompts: analyze_strategy, strategic_advisor, pulse (quick progress check-in), coach_me, build_workspace, scaling_up_sync, and more. These turn the agent into an active strategic collaborator, not just a query interface.
9 live resource URIs
Named resources that agents can read directly: stratafy://foundation, stratafy://strategies/tree, stratafy://initiatives, stratafy://metrics/summary, stratafy://insights, and more. Each resource provides structured, typed access to a specific layer of strategic data.
Permission-gated, role-aware access
Every tool and prompt is gated via role-based permissions. Roles include strategic_advisor, manager, editor, investor, and viewer — each receiving a tailored perspective. Agent token logging tracks every tool invocation, creating an audit trail of what AI agents accessed and acted on.