product: maestro audience: test-developer, operator, ai-assistant authority: normative

MCP Tools — Reports

Tools for retrieving historical test results, step data, execution logs, and artifacts.

All tool responses include a top-level station field containing the StationLabel configured for this server instance.

See also: tools-reference.md for the full tool index.

`search_reports`

Searches the test execution history. All filter parameters are optional and can be combined freely.

Parameter	Type	Required	Description
`serialNumber`	string	No	Filter by DUT serial number (exact match)
`testName`	string	No	Filter by test name (partial match)
`verdict`	string	No	Filter by verdict: `Pass`, `Fail`, `Abort`
`fromUtc`	string	No	ISO 8601 start of date range, e.g. `2025-01-01T00:00:00Z`
`toUtc`	string	No	ISO 8601 end of date range
`page`	int	No	Page number. Default: `1`
`pageSize`	int	No	Results per page. Default: `25`

Returns: results — items[] (summary rows), totalCount, page, pageSize

Calls: GET /api/test-results?...

`get_test_report_summary`

Returns the header-level details of a test execution: metadata, verdict, child executions, and artifact list. Step execution data is not included — use get_test_report_full or get_step_results for steps.

Parameter	Type	Required	Description
`executionId`	GUID	Yes	The execution ID

Returns: report — full TestResultDetailDto: testName, verdict, duration, tags, childExecutions, artifacts

Calls: GET /api/test-results/{id}

`get_test_report_full`

Returns a paginated page of step execution results for a test. Steps are capped at 100 per page by default. For tests with many steps, make multiple calls incrementing page until page >= steps.totalPages.

Parameter	Type	Required	Description
`executionId`	GUID	Yes	The execution ID
`page`	int	No	Page number, 1-indexed. Default: `1`
`pageSize`	int	No	Steps per page. Default: `100`

Returns: steps — items[], totalCount, page, pageSize, totalPages

Calls: GET /api/test-results/{id}/steps?page=&pageSize=

`get_step_results`

Returns paginated step execution results for a specific execution. Each step includes name, verdict, duration, and any measurement values recorded.

Parameter	Type	Required	Description
`executionId`	GUID	Yes	The execution ID
`page`	int	No	Page number. Default: `1`
`pageSize`	int	No	Steps per page. Default: `100`

Returns: steps — same shape as get_test_report_full

Calls: GET /api/test-results/{id}/steps?page=&pageSize=

`get_execution_logs`

Returns all log lines for a test execution with structured metadata (timestamp, step name, log level, source).

Parameter	Type	Required	Description
`executionId`	GUID	Yes	The execution ID

Returns: logs[] — each has id, stepExecutionId, stepName, logLevel, message, timestamp, source

Calls: GET /api/test-results/{id}/logs

For a flat unstructured list of log messages, use get_execution_log in tools-test-execution.md instead.

`list_artifacts`

Lists artifact metadata for a test execution. Artifacts are files produced by the test (e.g. oscilloscope screenshots, CSV exports).

Parameter	Type	Required	Description
`executionId`	GUID	Yes	The execution ID

Returns: artifacts[] — each has id, fileName, contentType, sizeBytes

Calls: GET /api/test-executions/{execId}/artifacts

`download_artifact`

Downloads the binary content of an artifact, returned as a base64-encoded string.

Parameter	Type	Required	Description
`executionId`	GUID	Yes	The execution ID that owns the artifact
`artifactId`	GUID	Yes	The artifact ID from `list_artifacts`

Returns: artifactId, sizeBytes, base64 (the artifact bytes, base64-encoded)

Calls: GET /api/test-executions/{execId}/artifacts/{artifactId}/download

For large artifacts, have the AI write the base64 string to a local file and decode it with [System.Convert]::FromBase64String(...) in PowerShell or base64 -d on Linux.