$ loading_
Query Copilot Studio telemetry and governance data in plain English.
The materials indicate this tool pulls Copilot Studio telemetry and governance data from Azure and Microsoft 365 APIs into a local SQLite database, implying local code execution, enterprise data access, and outbound network activity. Open-source MIT licensing is a positive factor, but sparse documentation, low adoption, and incomplete disclosure around endpoints/credentials make it a caution-level tool overall.
The materials say no keys or environment variables are required, but querying Azure and Microsoft 365 data typically relies on an existing Microsoft sign-in, CLI/auth context, or other ambient credentials. There is no explicit indication of secret harvesting, but reuse of local enterprise identity still warrants attention to scope and session misuse.
The description explicitly states that it pulls data from multiple Azure and Microsoft 365 APIs, so outbound network access and data transfer are inherent. Specific endpoints are not disclosed in the materials; there is no evidence of exfiltration to unrelated third parties, but the network behavior is incompletely documented and should be verified.
The system flags executes-code, and as an MCP tool it necessarily runs local logic for querying and syncing data. The materials do not show requests for unusual system privileges or unrelated high-risk operations, so this remains a normal caution-level local execution case.
The description says telemetry and governance data is pulled into a local SQLite database, which implies local persistence and possible handling of enterprise usage, governance, or audit information. The materials do not specify database location, retention, field sensitivity, or encryption; while no clear overprivilege is shown, it should be treated as having meaningful access to both local and enterprise data.
A positive factor is the public GitHub repository under MIT license, which materially reduces opaque closed-source risk. Caution remains because it comes from a third-party registry, shows 0 stars, has unknown maintenance status, and lacks a README, leaving limited verifiable information and requiring extra review of dependencies and implementation details.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Copilot Insights MCP Server" yet — see the docs or source repo.
Query Copilot Studio session volume, active users, and peak hours over the last 30 days, grouped by day.
A daily trend table with key peaks highlighted.
List governance risks in Copilot Studio related to permissions, connectors, or policies, sorted by severity.
A severity-sorted risk list with recommendations.
Find the Copilot Studio workflows or agents with the highest failure rates this week, and provide likely causes and related metrics.
High-failure items, related metrics, and likely cause analysis.
Check GitHub Copilot quota, rate limits, and reset times from MCP clients.
Query and manage Microsoft SQL Server databases with natural language.
An AI copilot for product development with search, SQL, and charts.
Search ServiceNow catalogs and place orders from Copilot Studio.
Query PostgreSQL with natural language, generate validated SQL, and optionally execute it.
Query normalized usage, cost, and dashboard data across OpenAI and Anthropic.