Chat with your context lake outside of Port
Why it matters
Engineering leaders and developers often need quick answers from their data like "Which services had the most incidents this quarter?", "What's our deployment frequency trend?", "Who owns this service and what's their on-call rotation?" but finding answers means switching between multiple tools and piecing information together manually. Natural language interaction lets anyone get answers instantly, grounded in your full engineering context.
What you can do
- Service & ownership queries: Ask about service ownership, dependencies, on-call rotations, and team structure.
- Metrics exploration: Query DORA metrics, delivery performance, and scorecard compliance across teams.
- Incident investigation: Investigate incidents with full context from connected tools like blast radius, recent deployments, related services.
- Cross-team comparison: Compare team performance across any metric in your context lake.
How Port helps
Through Port's MCP integration, you can connect AI tools like Claude, Cursor, or any MCP-compatible client to your software catalog & context lake. Ask questions in natural language and get answers grounded in your actual data, services, teams, deployments, incidents, metrics, and dependencies. MCP connectors extend this further by bringing real-time context from external tools like GitHub, Jira, Slack, PagerDuty, and more so the AI can answer questions that span your entire engineering stack, not just what's stored in the catalog.
Example scenario
An engineering director preparing for a quarterly review opens Claude and asks "Show me the top 5 teams by deployment frequency this quarter, and highlight any that have a change failure rate above 15%." In seconds, they get a formatted answer with the data pulled directly from Port no dashboard building, no SQL queries, no context-switching between tools. They follow up with "What incidents did the lowest-performing team have?" and get a summary pulled via the PagerDuty MCP connector in real time.