Skip to main content
Glama
NGS360 MCP Server

NGS360 MCP Server

Official
by NGS360

NGS360 MCP Server

A Model Context Protocol (MCP) server that exposes the NGS360 REST API as tools for AI assistants.

Overview

This MCP server wraps the NGS360 bioinformatics platform API, providing tools for:

Domain

Tools

Description

Auth

1 tool

Current-user profile (whoami)

Runs

14 tools

Sequencing run CRUD, sample sheets, metrics, demultiplexing

Jobs

6 tools

AWS Batch job submission, monitoring, and logs

Projects

11 tools

Project CRUD, samples, pipeline actions, vendor ingestion

Files

9 tools

File record management, S3 browsing, uploads, presigned downloads

Search

1 tool

Cross-index search (projects, runs, samples)

Workflows

11 tools

Workflow identity, versions, aliases, deployments

Pipelines

5 tools

Pipeline CRUD, workflow associations

QC Metrics

5 tools

QC record creation, search, retrieval

Vendors

5 tools

Vendor CRUD

Settings

3 tools

Application settings management

Platforms

3 tools

Workflow execution platform registry

Manifest

2 tools

Manifest retrieval and validation

WES

8 tools

GA4GH Workflow Execution Service (run/monitor workflows)

Related MCP server: AI Core MCP Server

Installation

cd mcp-server
pip install -e .

Or with uv:

cd mcp-server
uv pip install -e .

Configuration

The server uses environment variables for configuration:

Variable

Default

Description

NGS360_API_URL

http://localhost:8000

Base URL of the NGS360 API

NGS360_API_TOKEN

(empty)

Bearer token for NGS360 authentication

Usage

Running directly

ngs360-mcp-server

With Claude Desktop

Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "ngs360": {
      "command": "ngs360-mcp-server",
      "env": {
        "NGS360_API_URL": "https://your-ngs360-api.example.com",
        "NGS360_API_TOKEN": "your-bearer-token",
      }
    }
  }
}

With VS Code / Roo

Add to your MCP settings:

{
  "mcpServers": {
    "ngs360": {
      "command": "ngs360-mcp-server",
      "env": {
        "NGS360_API_URL": "https://your-ngs360-api.example.com",
        "NGS360_API_TOKEN": "your-bearer-token",
      }
    }
  }
}

Development

Project Structure

mcp-server/
├── pyproject.toml
├── README.md
└── ngs360_mcp_server/
    ├── __init__.py
    ├── client.py          # HTTP client wrapper for the NGS360 API
    ├── server.py          # MCP server entry point
    └── tools/
        ├── __init__.py
        ├── auth.py        # Auth / current-user tools
        ├── runs.py        # Sequencing run tools
        ├── jobs.py        # Batch job tools
        ├── projects.py    # Project & sample tools
        ├── files.py       # File management tools
        ├── search.py      # Cross-index search tools
        ├── workflows.py   # Workflow management tools
        ├── pipelines.py   # Pipeline management tools
        ├── qcmetrics.py   # QC metrics tools
        ├── vendors.py     # Vendor management tools
        ├── settings.py    # Settings tools
        ├── platforms.py   # Platform registry tools
        ├── manifest.py    # Manifest tools

Testing with MCP Inspector

npx @modelcontextprotocol/inspector ngs360-mcp-server

Adding New Tools

  1. Create or edit a file in ngs360_mcp_server/tools/

  2. Define a register_*_tools(mcp, client) function

  3. Register it in ngs360_mcp_server/server.py

Each tool is a decorated async function:

@mcp.tool()
async def my_new_tool(param1: str, param2: int = 10) -> dict:
    """Tool description shown to the AI assistant.

    Args:
        param1: Description of param1
        param2: Description of param2
    """
    return await client.get("/my/endpoint", params={"p": param1})

API Coverage

This MCP server covers the full NGS360 API surface at /api/v1/*:

  • GET /auth/me — Authenticated user profile (whoami)

  • POST/GET /runs — Create and list sequencing runs

  • GET/PUT /runs/{run_id} — Get/update a run

  • GET /runs/{run_id}/samplesheet — Sample sheet

  • GET /runs/{run_id}/metrics — Demux metrics

  • GET/POST /runs/demultiplex — Demux workflows

  • POST/GET/DELETE /runs/{run_id}/samples — Sample associations

  • POST/GET /jobs — Submit and list batch jobs

  • GET/PUT /jobs/{job_id} — Get/update job

  • GET /jobs/{job_id}/log — Job logs

  • POST/GET /projects — Create and list projects

  • GET/PUT/PATCH /projects/{project_id} — Project CRUD

  • POST/GET /projects/{project_id}/samples — Project samples

  • POST /projects/{project_id}/actions/submit — Pipeline submission

  • POST/GET /files — File record management

  • GET/PATCH/DELETE /files/{file_id} — File operations

  • GET /files/list — S3 browser

  • POST /files/upload — File upload

  • GET /files/download-url — Presigned download URL

  • GET /search — Global search

  • POST/GET /workflows — Workflow identity

  • POST/GET /workflows/{id}/versions — Versions

  • PUT/GET/DELETE /workflows/{id}/aliases — Aliases

  • POST/GET/DELETE /workflows/.../deployments — Deployments

  • POST/GET /pipelines — Pipeline CRUD

  • POST/DELETE /pipelines/{id}/workflows — Workflow membership

  • POST/GET/DELETE /qcmetrics — QC records

  • GET/POST /qcmetrics/search — QC search

  • POST/GET/PUT/DELETE /vendors — Vendor management

  • GET/PUT /settings — Application settings

  • POST/GET /platforms — Platform registry

  • GET/POST /manifest — Manifest operations

GA4GH WES API (/ga4gh/wes/v1/*)

  • GET /service-info — WES service capabilities

  • GET /runs — List workflow runs

  • POST /runs — Submit a workflow run

  • GET /runs/{run_id} — Full run log

  • GET /runs/{run_id}/status — Run state

  • POST /runs/{run_id}/cancel — Cancel a run

  • GET /runs/{run_id}/tasks — List tasks for a run

  • GET /runs/{run_id}/tasks/{task_id} — Get task details

Available Tools

82 tools
add_sample_to_projectC

Add a sample to a project.

Args: project_id: The project identifier sample_id: Sample identifier string attributes: Optional sample attributes run_id: Optional run_id to associate the sample with

Inputs: sample_id (required), project_id (required), run_id, attributes.

add_workflow_to_pipelineC

Add a workflow to a pipeline.

Args: pipeline_id: The pipeline UUID workflow_id: UUID of the workflow to associate

Inputs: pipeline_id (required), workflow_id (required).

associate_sample_with_runB

Associate a sample with a sequencing run.

Args: run_id: The run identifier sample_id: UUID of the sample to associate

Inputs: run_id (required), sample_id (required).

browse_s3A

Browse files and folders at an S3 URI.

Args: uri: S3 URI to list (e.g., s3://bucket/folder/)

Inputs: uri (required).

clear_samples_for_runA

Remove all sample associations and cleanup for a run (re-demux scenario).

Args: run_id: The run identifier

Inputs: run_id (required).

create_file_recordB

Create a new file record (external reference).

At least one entity association is required.

Args: uri: File location (s3://, file://, etc.) project_id: Project business key (string) sequencing_run_id: SequencingRun UUID qcrecord_id: QCRecord UUID pipeline_id: Pipeline UUID original_filename: Original filename before renaming size: File size in bytes source: Origin of file record created_by: User identifier storage_backend: Storage type (LOCAL, S3, AZURE, GCS) samples: Sample associations [{"sample_name": "S1", "role": "tumor"}] hashes: Hash values {"md5": "abc...", "sha256": "def..."} tags: Key-value metadata {"type": "alignment", "format": "bam"}

Inputs: uri (required), size, tags, hashes, +9 more.

create_pipelineB

Create a new pipeline (named collection of workflows).

Args: name: Pipeline name version: Optional version string attributes: Optional key-value attributes workflow_ids: Optional list of workflow UUIDs to include

Inputs: name (required), version, attributes, workflow_ids.

create_platformB

Create a new workflow execution platform.

Args: name: Platform name (e.g., Arvados, SevenBridges)

Inputs: name (required).

create_projectB

Create a new project.

Args: name: Project name attributes: Optional list of key-value attributes [{"key": "type", "value": "RNA-Seq"}]

Inputs: name (required), attributes.

create_qcrecordA

Create a new QC record with metrics and output files.

Exactly one of project_id or sequencing_run_id must be provided.

Args: project_id: Project-scoped record (e.g., P-1234) sequencing_run_id: Run-scoped record (run_id string) workflow_run_id: Optional workflow run UUID (provenance) metadata: Pipeline metadata {"pipeline": "RNA-Seq", "version": "2.0"} metrics: List of metric groups, each with: - name: Metric group name - samples: Optional [{"sample_name": "S1", "role": "tumor"}] - values: {"reads": 50000000, "alignment_rate": 95.5} output_files: Optional list of file records to create

Inputs: metrics, metadata, project_id, output_files, +2 more.

create_runC

Create a new sequencing run.

Args: run_id: Unique run identifier run_date: Run date in YYYY-MM-DD format machine_id: Instrument machine ID run_number: Run number (string) flowcell_id: Flowcell identifier experiment_name: Optional experiment name run_folder_uri: Optional URI to the run folder status: Optional status (In Progress, Uploading, Ready, Resync) run_time: Optional run time in HHMM format (4 digits)

Inputs: run_id (required), run_date (required), machine_id (required), run_number (required), +5 more.

create_vendorD

Create a new vendor.

Args: vendor_id: Unique vendor identifier name: Vendor display name description: Vendor description bucket: Optional S3 bucket associated with vendor

Inputs: name (required), vendor_id (required), description (required), bucket.

create_workflowB

Create a new workflow identity.

Args: name: Workflow name attributes: Optional key-value attributes [{"key": "category", "value": "alignment"}]

Inputs: name (required), attributes.

create_workflow_deploymentA

Deploy a workflow version on a specific platform.

Args: workflow_id: The workflow UUID version_id: The version UUID engine: Platform name (e.g., Arvados, SevenBridges) external_id: Workflow identifier on the external platform

Inputs: engine (required), version_id (required), external_id (required), workflow_id (required).

create_workflow_versionA

Create a new version for a workflow.

Version number is auto-incremented.

Args: workflow_id: The workflow UUID definition_uri: URI to the workflow definition file

Inputs: workflow_id (required), definition_uri (required).

delete_fileA

Delete a file record and all associated data (superuser only, irreversible).

Args: file_id: The file UUID

Inputs: file_id (required). Returns: result

delete_qcrecordA

Delete a QC record and all associated data (irreversible).

Args: qcrecord_id: The QC record UUID

Inputs: qcrecord_id (required).

delete_sample_from_projectA

Delete a sample from a project (superuser only, irreversible).

Args: project_id: The project identifier sample_id: The sample identifier to delete

Inputs: sample_id (required), project_id (required). Returns: result

delete_vendorC

Delete a vendor.

Args: vendor_id: The vendor identifier

Inputs: vendor_id (required). Returns: result

delete_workflow_aliasC

Remove an alias from a workflow.

Args: workflow_id: The workflow UUID alias: Alias name to remove

Inputs: alias (required), workflow_id (required). Returns: result

delete_workflow_deploymentC

Remove a platform deployment.

Args: workflow_id: The workflow UUID version_id: The version UUID deployment_id: The deployment UUID

Inputs: version_id (required), workflow_id (required), deployment_id (required). Returns: result

get_demux_workflow_configB

Get configuration for a specific demultiplex workflow.

Args: workflow_id: The workflow identifier run_id: Optional run ID to prepopulate s3_run_folder_path

Inputs: workflow_id (required), run_id.

get_fileB

Get file metadata by UUID.

Args: file_id: The file UUID

Inputs: file_id (required).

get_file_versionsB

Get all versions of a file.

Args: file_id: The file UUID (any version)

Inputs: file_id (required).

get_jobC

Get details of a specific batch job.

Args: job_id: The job UUID

Inputs: job_id (required).

get_job_logC

Get the full log output for a batch job.

Args: job_id: The job UUID

Inputs: job_id (required).

get_job_log_paginatedA

Get paginated log output for a batch job.

Args: job_id: The job UUID limit: Number of log lines to return (1-10000) next_token: Pagination token from previous response start_from_head: Start from beginning (true) or end (false)

Inputs: job_id (required), limit, next_token, start_from_head.

get_latest_manifestA

Get the latest manifest file path from an S3 bucket.

Searches recursively for CSV files containing 'manifest' in their name.

Args: s3_path: S3 path to search (e.g., s3://bucket-name/path/to/manifests)

Inputs: s3_path (required). Returns: result

get_pipelineB

Get details of a specific pipeline.

Args: pipeline_id: The pipeline UUID

Inputs: pipeline_id (required).

get_platformA

Get a platform by name.

Args: name: Platform name (e.g., Arvados, SevenBridges)

Inputs: name (required).

get_projectB

Get details of a specific project.

Args: project_id: The project identifier (e.g., 'P-1234')

Inputs: project_id (required).

get_project_samplesC

Get samples for a project with pagination.

Args: project_id: The project identifier skip: Number of records to skip limit: Maximum number of records to return sort_by: Field to sort by sort_order: Sort order (asc or desc) include_files: Whether to include file data for each sample

Inputs: project_id (required), skip, limit, sort_by, +2 more.

get_qcrecordB

Get a specific QC record by UUID.

Args: qcrecord_id: The QC record UUID

Inputs: qcrecord_id (required).

get_runB

Get details of a specific sequencing run.

Args: run_id: The run identifier (e.g., '240101_A00000_0001_FLOWCELLID')

Inputs: run_id (required).

get_run_metricsC

Get demultiplexing metrics for a sequencing run.

Args: run_id: The run identifier

Inputs: run_id (required).

get_run_samplesheetC

Get the sample sheet for a sequencing run.

Args: run_id: The run identifier

Inputs: run_id (required).

get_samples_for_runB

List sample associations for a sequencing run.

Args: run_id: The run identifier

Inputs: run_id (required).

get_settingC

Get a specific setting by key.

Args: key: The setting key identifier

Inputs: key (required).

get_settings_by_tagA

Get settings filtered by a tag key-value pair.

Args: tag_key: Tag key to filter by (e.g., 'category') tag_value: Tag value to filter by (e.g., 'storage')

Inputs: tag_key (required), tag_value (required).

get_vendorC

Get details of a specific vendor.

Args: vendor_id: The vendor identifier

Inputs: vendor_id (required).

get_workflowC

Get details of a specific workflow.

Args: workflow_id: The workflow UUID

Inputs: workflow_id (required).

get_workflow_versionA

Get a specific workflow version.

Args: workflow_id: The workflow UUID version_id: The version UUID

Inputs: version_id (required), workflow_id (required).

ingest_vendor_dataC

Ingest vendor data for a project.

Args: project_id: The project identifier files_uri: Source S3 bucket/prefix of data to ingest manifest_uri: S3 path to the vendor manifest

Inputs: files_uri (required), project_id (required), manifest_uri (required).

list_demux_workflowsB

List available demultiplex workflows.

Inputs: no named parameters.

list_filesA

List or search files.

Args: uri: Filter by URI (returns latest version) entity_type: Filter by entity type (PROJECT, SEQUENCING_RUN, SAMPLE, QCRECORD, PIPELINE) entity_id: Filter by entity ID (requires entity_type) include_archived: Include archived files page: Page number per_page: Items per page

Inputs: uri, page, per_page, entity_id, +2 more.

list_jobsA

List batch jobs with optional filtering and sorting.

Args: skip: Number of records to skip limit: Maximum number of records to return user: Optional filter by username status_filter: Optional filter by status (SUBMITTED, PENDING, RUNNABLE, STARTING, RUNNING, SUCCEEDED, FAILED) sort_by: Field to sort by (defaults to submitted_on) sort_order: Sort order (asc or desc)

Inputs: skip, user, limit, sort_by, +2 more.

list_pipelinesB

List pipelines with pagination and sorting.

Args: page: Page number (1-indexed) per_page: Number of items per page sort_by: Field to sort by sort_order: Sort order (asc or desc)

Inputs: page, sort_by, per_page, sort_order.

list_platformsA

List all registered workflow execution platforms.

Inputs: no named parameters.

list_projectsB

List projects with pagination and sorting.

Args: page: Page number (1-indexed) per_page: Number of items per page sort_by: Field to sort by sort_order: Sort order (asc or desc)

Inputs: page, sort_by, per_page, sort_order.

list_runsC

List sequencing runs with pagination and sorting.

Args: page: Page number (1-indexed) per_page: Number of items per page sort_by: Field to sort by sort_order: Sort order (asc or desc)

Inputs: page, sort_by, per_page, sort_order.

list_vendorsA

List vendors with pagination and sorting.

Args: skip: Number of records to skip limit: Maximum number of records to return sort_by: Field to sort by sort_order: Sort order (asc or desc)

Inputs: skip, limit, sort_by, sort_order.

list_workflow_aliasesA

List aliases for a workflow.

Args: workflow_id: The workflow UUID alias: Optional filter to a specific alias name

Inputs: workflow_id (required), alias.

list_workflow_deploymentsA

List deployments for a workflow.

Can list across all versions or for a specific version.

Args: workflow_id: The workflow UUID version_id: Optional version UUID (if omitted, lists across all versions) alias: Optional alias filter (resolves to version) engine: Optional engine/platform filter

Inputs: workflow_id (required), alias, engine, version_id.

list_workflowsC

List workflows with pagination and sorting.

Args: page: Page number (1-indexed) per_page: Number of items per page sort_by: Field to sort by sort_order: Sort order (asc or desc)

Inputs: page, sort_by, per_page, sort_order.

list_workflow_versionsA

List all versions of a workflow.

Args: workflow_id: The workflow UUID

Inputs: workflow_id (required).

patch_projectA

Partially update a project (merge/upsert semantics).

Unlike update_project, this does NOT remove attributes absent from the request.

Args: project_id: The project identifier name: Optional new name attributes: Optional attributes to upsert

Inputs: project_id (required), name, attributes.

remove_sample_from_runC

Remove a single sample association from a run.

Args: run_id: The run identifier sample_id: The sample identifier to remove

Inputs: run_id (required), sample_id (required). Returns: result

remove_workflow_from_pipelineA

Remove a workflow from a pipeline.

Args: pipeline_id: The pipeline UUID workflow_id: The workflow UUID to remove

Inputs: pipeline_id (required), workflow_id (required). Returns: result

search_projectsB

Search projects by project_id or name.

Args: query: Search query string page: Page number (1-indexed) per_page: Number of items per page sort_by: Field to sort by (project_id or name) sort_order: Sort order (asc or desc)

Inputs: query (required), page, sort_by, per_page, +1 more.

search_qcrecordsB

Search QC records with filters.

Args: project_id: Filter by project ID sequencing_run_id: Filter by sequencing run_id string workflow_run_id: Filter by workflow run UUID latest: Return only newest record per scope (default True) page: Page number per_page: Results per page

Inputs: page, latest, per_page, project_id, +2 more.

search_qcrecords_advancedA

Search QC records with advanced filtering (POST).

Args: filter_on: Filter dictionary, e.g.: {"project_id": "P-1234", "metadata": {"pipeline": "RNA-Seq"}} page: Page number per_page: Results per page latest: Return only newest record per scope

Inputs: page, latest, per_page, filter_on.

search_runsC

Search sequencing runs using OpenSearch.

Args: query: Search query string page: Page number (1-indexed) per_page: Number of items per page sort_by: Field to sort by (run_id or experiment_name) sort_order: Sort order (asc or desc)

Inputs: query (required), page, sort_by, per_page, +1 more.

set_workflow_aliasA

Set or update a workflow version alias (e.g., 'production').

Args: workflow_id: The workflow UUID alias: Alias name (e.g., production, development) workflow_version_id: UUID of the version to point to

Inputs: alias (required), workflow_id (required), workflow_version_id (required).

submit_demux_jobC

Submit a demultiplex workflow job.

Args: workflow_id: The workflow identifier run_id: The run to demultiplex inputs: Workflow input parameters

Inputs: inputs (required), run_id (required), workflow_id (required).

submit_jobB

Submit a new batch job to AWS Batch.

Args: job_name: Name for the batch job job_definition: AWS Batch job definition ARN job_queue: AWS Batch job queue name command: Command string to execute user: Username submitting the job environment: Optional list of environment variables [{"name": "KEY", "value": "VAL"}]

Inputs: user (required), command (required), job_name (required), job_queue (required), +2 more.

submit_pipeline_jobB

Submit a pipeline job to AWS Batch for a project.

Args: project_id: The project identifier action: Pipeline action (create-project or export-project-results) platform: Platform name (Arvados or SevenBridges) project_type: Pipeline workflow type (e.g., RNA-Seq, WGS) reference: Export reference (required for export action) auto_release: Auto-release flag for export action

Inputs: action (required), platform (required), project_id (required), project_type (required), +2 more.

update_fileA

Update a file record (superuser only).

Only provided fields are updated.

Args: file_id: The file UUID uri: New URI original_filename: New original filename size: New file size source: New source created_by: New creator storage_backend: New storage backend

Inputs: file_id (required), uri, size, source, +3 more.

update_jobB

Update a batch job's status or metadata.

Args: job_id: The job UUID log_stream_name: Optional CloudWatch log stream name status: Optional new status (SUBMITTED, PENDING, RUNNABLE, STARTING, RUNNING, SUCCEEDED, FAILED) viewed: Optional viewed flag

Inputs: job_id (required), status, viewed, log_stream_name.

update_projectA

Full replacement update of a project (PUT semantics).

Attributes provided here replace all existing attributes.

Args: project_id: The project identifier name: Optional new name attributes: Optional list of attributes (replaces all existing)

Inputs: project_id (required), name, attributes.

update_run_statusA

Update the status of a sequencing run.

Args: run_id: The run identifier run_status: New status (In Progress, Uploading, Ready, Resync)

Inputs: run_id (required), run_status (required).

update_settingB

Update a setting (value, name, description, tags).

The key itself cannot be changed.

Args: key: The setting key value: Optional new value name: Optional new display name description: Optional new description tags: Optional new tags list [{"key": "category", "value": "storage"}]

Inputs: key (required), name, tags, value, +1 more.

update_vendorC

Update a vendor's information.

Args: vendor_id: The vendor identifier name: Optional new name description: Optional new description bucket: Optional new bucket

Inputs: vendor_id (required), name, bucket, description.

validate_manifestA

Validate a manifest CSV file from S3.

Checks required fields, data format, value constraints, and file existence.

Args: manifest_uri: S3/GS path to the manifest CSV file files_uri: S3/GS path where files in manifest are located manifest_version: Manifest version to validate against (e.g., 'DTS12.1') post_to_api: If true, post samples to API after successful validation

Inputs: manifest_uri (required), files_uri, post_to_api, manifest_version.

wes_cancel_runB

Cancel a running workflow.

Args: run_id: The workflow run identifier to cancel

Inputs: run_id (required).

wes_get_run_logC

Get detailed information about a workflow run.

Returns the run request, state, run log (stdout/stderr), task logs, and output files.

Args: run_id: The workflow run identifier

Inputs: run_id (required).

wes_get_run_statusA

Get the current status of a workflow run.

Returns an abbreviated status with the run_id and state (UNKNOWN, QUEUED, INITIALIZING, RUNNING, PAUSED, COMPLETE, EXECUTOR_ERROR, SYSTEM_ERROR, CANCELED, CANCELING, PREEMPTED).

Args: run_id: The workflow run identifier

Inputs: run_id (required).

wes_get_service_infoA

Get information about the WES service.

Returns supported workflow types, WES versions, filesystem protocols, workflow engine versions, default parameters, and system state counts.

Inputs: no named parameters.

wes_get_taskA

Get details of a specific task within a workflow run.

Returns log information including name, command, start/end time, stdout, stderr, exit code, and system logs.

Args: run_id: The workflow run identifier task_id: The task identifier

Inputs: run_id (required), task_id (required).

wes_list_runsA

List workflow runs from the WES service.

Returns a paginated list of workflow runs the caller has permission to see.

Args: page_size: Number of runs to return per page page_token: Token for retrieving the next page of results

Inputs: page_size, page_token.

wes_list_tasksA

List tasks for a workflow run.

Returns a paginated list of tasks executed as part of the workflow run.

Args: run_id: The workflow run identifier page_size: Number of tasks to return per page page_token: Token for retrieving the next page of results

Inputs: run_id (required), page_size, page_token.

wes_run_workflowB

Submit a new workflow run to the WES service.

Args: workflow_url: URL or relative path to the workflow definition (CWL/WDL) workflow_type: Workflow language type (e.g., "CWL", "WDL") workflow_type_version: Version of the workflow language (e.g., "v1.0") workflow_params: Input parameters for the workflow (JSON object) workflow_engine: Workflow engine to use (e.g., "cromwell", "toil") workflow_engine_version: Version of the workflow engine workflow_engine_parameters: Additional engine-specific parameters tags: Arbitrary key-value tags for the run

Inputs: workflow_url (required), workflow_type (required), workflow_type_version (required), tags, +4 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    C2.7/5.0

    Scored across 82 tools

    Disambiguation3/5

    Many tools are distinct by entity and action (e.g., create_project vs create_run), but there is overlap between search variants (search, search_projects, search_runs, etc.) and submit job variants (submit_job, submit_demux_job, submit_pipeline_job). Some tools like get_run_metrics and get_samples_for_run could be confused with run-related queries. Overall, most tools are distinguishable but with some ambiguity.

    Naming Consistency4/5

    Tool names consistently use verb_noun snake_case (e.g., add_sample_to_project, create_workflow_version). WES tools have a clear wes_ prefix. Minor inconsistencies exist, such as 'clear_samples_for_run' vs 'remove_sample_from_run' and 'update_run_status' vs 'update_job', but these are still clear. No mixing of conventions.

    Tool Count2/5

    82 tools is excessive for most domains, suggesting overly fine-grained operations and redundancy. For example, there are 5 search tools, 5 job submission tools, and many CRUD tools for each entity. This likely overwhelms agents and indicates poor scoping. A well-scoped server would have 15-25 tools.

    Completeness3/5

    The server covers a broad range of NGS entities (projects, samples, runs, workflows, pipelines, QC, files, vendors, WES). However, some entities lack full CRUD (e.g., no delete_pipeline, delete_workflow, update_sample). The search and job submission areas are overcovered, while others like sample updates are missing. Notable gaps exist.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers