A technical project manager finds Copilot answers about release readiness are generic because it cannot see Jira tickets, Confluence specs, and Postgres metrics. The team requires live governed context without pasting secrets into chat. Which integration approach should she choose to connect agentic workflows to MCP servers to improve context?
Select an answer to reveal the explanation.
Short Explanation
You can give your agent fresh, governed release context without pasting secrets into chat. MCP servers let it use live tools across Jira, docs, and database metrics.
Full Explanation
Connecting governed MCP servers for Jira, docs, and the database is the right approach because agents can call authenticated, scoped tools to combine ticket status, specification details, and current metrics for release-readiness answers. This supplies live context without pasting secrets into chat while supporting least privilege and auditability. Connecting only Jira and Confluence leaves out the database metrics the team needs, so the agent cannot assess release readiness across all required sources. Nightly cached context can become stale when tickets change, specs are edited, or metrics shift during the day, so it does not meet the need for live context. Live servers that supply only read-only prompt text provide context but not callable tools, limiting the agent’s ability to query and work with source data. An exam caveat is that governed access alone is not enough: the approach must also be live, provide tools, and cover Jira, docs, and the database. As a practical check, verify access scopes, secret handling, and audit logs, then test whether the agent can connect open blockers to relevant spec sections and current database metrics.