# masud-dot/building-mcp-servers-with-python

Source: https://glimind.com/server/github%2Fmasud-dot%2Fbuilding-mcp-servers-with-python
Registry id: `github/masud-dot/building-mcp-servers-with-python`
Status: unknown

## Tools (31)

- `acknowledge_incident`
- `bulk_create_tasks`: Create several tasks at once and return them. Reports progress as it goes, so a host can show how far through it has got.
- `check_endpoint`: Fetch a URL and report what came back. VULNERABLE: the caller chooses the host, and redirects are followed.
- `check_stock`: How many copies of a book are in stock.
- `complete_task`: Mark a task as done and return its new state. Completing an already-completed task is not an error.
- `create_note`: Record a short note and return it.
- `create_task`: Add a task to the list and return a confirmation.
- `delete_pipeline_run`
- `describe_table`: Return the columns and types of one table. Call this before sample_table so you know what the columns are.
- `get_by_handle`: Retrieve work started by scan_by_handle.
- `get_deployment_history`
- `get_scan_result`: Retrieve a log scan by its handle.
- `get_service_health`: Current health of one service, from monitoring. Live data, so it changes between calls.
- `get_task`: Return one task by its identifier.
- `get_test_run`: Return the summary of one test run.
- `list_failures`: Return the failing tests of one run. Stack traces are trimmed. When trace_truncated is true, the full trace is longer than what is shown here.
- `list_services`: List every service this server knows about. Start here. Other tools take a service name from this list and refuse anything else.
- `list_tables`: List the tables this server can read. Only these tables are available. Anything else is refused, including tables that exist in the database.
- `list_tasks`
- `list_test_runs`: List recent test runs, newest first. Returns run identifiers that get_test_run and list_failures accept.
- `model`: Show the current tasks. Renders a board in hosts that support MCP Apps, and returns the same information as text in hosts that do not.
- `query_incidents`
- `run_diagnostic`: Run one named diagnostic and return its output. Only diagnostics defined in this server can run. There is no way to pass a command or an argument.
- `run_tests`: Run a test suite and return its output. VULNERABLE: `target` is interpolated into a shell string.
- `sample`: Return a fixed reading for the given label.
- `sample_table`
- `scan_as_task`: Long work. Poll with the extension's tasks/get. The tool cannot declare `execution.task_support`: the field exists in mcp_types.Tool, and MCPServer.add_tool has
- `scan_by_handle`: Medium work. Returns a handle; poll get_by_handle. Note the parameterised return type. A bare `dict` produces no output schema and no structured content at all.
- `scan_with_progress`: Short work. Reports progress and blocks until done.
- `start_log_scan`: Start a log scan and return a handle. The scan outlives this call. Any instance of this server can answer get_scan_result.
- `start_quality_scan`: Start a data-quality scan and return a handle. The scan may outlive this call. Come back with get_scan_result, which any instance of this server can answer.

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.