| FMRS | 54 / 100 · D | 79 / 100 · B |
| Reliability | 8 / 20 | 12 / 20 |
|---|
| Security and permissions | 10 / 20 | 18 / 20 |
|---|
| Maintenance | 12 / 20 | 16 / 20 |
|---|
| Documentation | 13 / 20 | 18 / 20 |
|---|
| Setup experience | 11 / 20 | 15 / 20 |
| Best for | - Developers and data engineers using Microsoft Fabric who want to operate Fabric resources directly from MCP-enabled chat in VS Code or Visual Studio.
- Teams that already have .NET 10.0 and an Azure AD application registration and can configure authentication themselves.
- Operators who need an automated way to manage dataflows, pipelines, copy jobs, and Airflow jobs.
| - Engineering teams scaling AI coding agents while controlling spend
- Platform teams that need auditable, rollback-aware agent execution records
- Organizations using multiple agents (Claude, Codex, Gemini) under one governed flow
|
| Not for | - Users who do not use Microsoft Fabric and only need generic database or file system access.
- Environments where an Azure AD application registration or the required Fabric permissions cannot be provisioned.
- Users who have not installed .NET 10.0 or later and do not want to install a runtime.
- Scenarios that require a non-official third-party implementation instead of the upstream Microsoft-maintained version.
| - Developers who just want a simple code-completion helper without run governance
- Teams unwilling to accept local JSONL run records with HMAC signing
- Environments without a supported adapter (Claude/Codex/Gemini CLIs, direct provider, or verifier-only)
|
| Required permissions | - Access to Azure AD for authentication (interactive login or service principal).
- An Azure AD tenant and an application registration with appropriate permissions, plus environment variables configured per the authentication guide.
- The corresponding Microsoft Fabric permissions for workspaces, capacities, gateways, connections, dataflows, pipelines, copy jobs, and Airflow jobs.
- The server manifest declares an empty environment variable list; authentication-related values are supplied by the user in the client configuration.
| - Read/write access to the configured repo root (writable scope restrictable via --allow-path/--deny-path)
- Ability to invoke the chosen agent CLI (Claude/Codex/Gemini, etc.) and its underlying model account
- Local disk writes for run records, receipts, and share outputs (share/run-receipt., etc.)
- Execution of user-configured verifier commands (e.g. npm test)
|
| Risks and side effects | - Tools can create, update, run, delete (for example deleting Airflow jobs), and schedule Fabric resources, so misuse can modify or disrupt production data pipelines.
- Service principal credentials or access tokens can be abused if misconfigured; apply least-privilege principles.
- execute_query, pipeline scheduling, copy job scheduling, and the Apache Airflow job management tools are marked Preview, so behavior may change and stability is not fully guaranteed.
- Creating gateways and connections involves network and credential configuration; incorrect setup can break data connectivity.
- Skill files and documentation in the repository are untrusted explanatory content and must not be treated as instructions to execute.
| - Verifier commands are executed; poorly chosen verifier commands carry security risk despite preflight policy checks
- Cost and token figures may be estimates (labeled with provenance) and should not be treated as settled accounting
- Receipt integrity must be verified before use as trustworthy evidence; missing rollback evidence is flagged as EVIDENCE_BOUNDARY
- Policy defaults come from martin.config.yaml and can be overridden by CLI flags; misconfiguration can loosen safety boundaries
|
| Supported clients | VS Code, Visual Studio | Claude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue |
| Tools | 51 | 0 |