Skip to main content
Glama
arunkumars-mf

z/OS FTP MCP Server

by arunkumars-mf
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

z/OS FTP MCP Server

PyPI version PyPI Downloads Python versions License: MIT

A Model Context Protocol (MCP) server for interacting with z/OS mainframe systems via FTP.

Features

  • List datasets from mainframe catalog with pattern matching

  • Download datasets in binary and text formats with encoding conversion

  • Download members from partitioned datasets (PDS) with parallel support

  • Submit and monitor JCL jobs

  • Advanced text processing: encoding conversion, line ending control, trailing spaces handling

  • Configurable write protection for safe production use

  • Connection management via environment variables

Related MCP server: ibmi-mcp-server

Installation

For MCP Server Usage

No installation needed! The MCP configuration with uvx will automatically download and run the package.

For Python Library Usage

pip install zos-ftp-mcp

Configuration

Set environment variables for connection:

export ZFTP_HOST="your-mainframe-host"
export ZFTP_PORT="21"
export ZFTP_USER="your-username"
export ZFTP_PASSWORD="your-password"
export ZFTP_TIMEOUT="600.0"
export ZFTP_DOWNLOAD_PATH="/path/to/downloads"
export ZFTP_DEBUG="false"  # Set to "true" for detailed FTP protocol logging
export ZFTP_ALLOW_WRITE="false"  # Set to "true" to enable upload/submit operations

Advanced Text Processing

Control how text files are downloaded from EBCDIC mainframe to ASCII/UTF-8:

# Character encoding conversion (EBCDIC to ASCII/UTF-8)
export ZFTP_DEFAULT_ENCODING="IBM-037,UTF-8"  # US/Canada EBCDIC to UTF-8
# Other options: IBM-1047,UTF-8 (Latin-1), IBM-285,UTF-8 (UK)

# Line ending format
export ZFTP_DEFAULT_LINE_ENDING="LF"  # Unix/Linux style
# Options: CRLF (Windows), LF (Unix), CR (old Mac), NONE

# Preserve trailing spaces in fixed-length records
export ZFTP_PRESERVE_TRAILING_SPACES="false"  # Strip trailing spaces
# Set to "true" to keep 80-character fixed-length records intact

When to use these settings:

  • Encoding: Essential for proper character conversion (JCL, COBOL, logs)

  • Line endings: Cross-platform compatibility, version control (Git)

  • Trailing spaces: Keep "false" for most text files, "true" for fixed-format data

Write Operations

By default, the MCP server is read-only for safety. To enable write operations:

export ZFTP_ALLOW_WRITE="true"

When enabled, you can:

  • Upload files to mainframe datasets (upload_dataset)

  • Submit JCL jobs (submit_job)

Use with caution in production environments. Read-only operations (list, download, monitor jobs) are always available.

Usage

As MCP Server

{
  "mcpServers": {
    "zos-ftp-mcp": {
      "command": "uvx",
      "args": ["zos-ftp-mcp"],
      "env": {
        "ZFTP_HOST": "your-mainframe-host",
        "ZFTP_USER": "your-username",
        "ZFTP_PASSWORD": "your-password",
        "ZFTP_DOWNLOAD_PATH": "/path/to/downloads"
      }
    }
  }
}

Direct Command Line

zos-ftp-mcp

Tools Available

Dataset Operations

  • list_catalog(pattern, limit, offset) - List datasets matching pattern with pagination

  • download_binary(source_dataset, target_file) - Download dataset in binary mode

  • download_text(source_dataset, target_file, encoding, line_ending, preserve_trailing_spaces) - Download with text processing

  • download_pds_members(dataset, target_dir, members, encoding, line_ending, ...) - Download PDS members with advanced options

  • upload_dataset(source_file, target_dataset, binary, lrecl, blksize, recfm, space) - Upload file to dataset (requires ZFTP_ALLOW_WRITE=true)

  • get_vsam_info(dataset, limit, offset) - Get VSAM dataset information with pagination (requires ZFTP_ALLOW_WRITE=true)

  • get_gdg_info(gdg_base, limit, offset) - Get GDG base attributes and generations with pagination (requires ZFTP_ALLOW_WRITE=true)

Job Operations

  • submit_job(jcl, jobname) - Submit JCL job (requires ZFTP_ALLOW_WRITE=true)

  • list_jes_jobs(jobmask, owner, status, limit, offset) - List jobs with filtering and pagination

  • get_job_info(jobid) - Get job details

  • download_job_spool(jobid, target_file) - Download job output with return code extraction

Connection Management

  • get_connection_info() - Show current connection settings

Sample Usage Prompts

Once configured, you can use these prompts:

Dataset Discovery

"List all datasets starting with SYS1 to explore system datasets"

"Show me all user datasets matching MYUSER.* pattern"

Data Download

"Download the dataset MYUSER.COBOL.SOURCE to my local downloads folder with UTF-8 encoding"

"Download all members from the PDS MYUSER.COBOL.COPYLIB to a local directory"

Data Upload (requires ZFTP_ALLOW_WRITE=true)

"Upload my local file test.jcl to dataset MYUSER.JCL.TEST with lrecl=80 and recfm=FB"

"Upload local file member.cbl to PDS member MYUSER.COBOL.SOURCE(TESTPGM)"

Job Operations

"Submit this JCL job and wait for completion" (requires ZFTP_ALLOW_WRITE=true)

"List all my jobs that are in OUTPUT status"

"Download the spool output for job JOB12345 and show me the return code"

Connection Management

"Show me the current FTP connection settings"

Dependencies

  • mcp - Model Context Protocol

  • Python 3.10+ standard library (ftplib, pathlib, etc.)

Advanced Features

Pagination Support

To avoid overwhelming context windows with large result sets, several tools support pagination:

Tools with pagination:

  • list_catalog - Paginate through datasets

  • list_jes_jobs - Paginate through jobs

  • get_vsam_info - Paginate through VSAM datasets

  • get_gdg_info - Paginate through GDG generations

Usage:

# Get first 50 datasets
list_catalog('USER.*', limit=50)

# Get next 50 datasets
list_catalog('USER.*', limit=50, offset=50)

# Get first 20 jobs
list_jes_jobs(status='OUTPUT', limit=20)

# Get first 10 GDG generations
get_gdg_info('AWS.M2.CARDDEMO.TRANSACT.BKUP', limit=10)

Response includes:

  • count - Number of items returned in this page

  • total - Total number of items available

  • offset - Current offset

  • has_more - True if more items are available

JESINTERFACELEVEL Limitations

The server automatically detects your mainframe's JES interface level. With JESINTERFACELEVEL=1 (common in older systems):

Job Submission:

  • Jobname in JCL must be userid + exactly one character

  • Example: If userid is USER123, jobname must be USER123J

  • The server validates this automatically and provides clear error messages

Job Listing:

  • Server-side filtering may not work

  • Client-side filtering is applied automatically

  • Jobs may be purged quickly from spool

Workaround: The server handles these limitations transparently. Just be aware that job retrieval may fail if jobs are purged quickly.

Parallel PDS Downloads

Download PDS members faster using multiple FTP connections:

# Sequential (default)
download_pds_members(dataset, target_dir, members, ftp_threads=1)

# Parallel (2-16 threads)
download_pds_members(dataset, target_dir, members, ftp_threads=4)

Parallel downloads are significantly faster for large PDSs (>10 members).

Per-Operation Overrides

While defaults are set via environment variables, you can override them per operation:

# Use different line ending for specific file
download_text(dataset, target_file, line_ending='CRLF')

# Preserve trailing spaces for fixed-format data
download_text(dataset, target_file, preserve_trailing_spaces=True)

# Use different encoding
download_pds_members(dataset, target_dir, members, encoding='IBM-1047,UTF-8')

License

MIT License - see LICENSE file for details.

Available Tools

8 tools
download_binaryB

Download a dataset from mainframe in binary format

ParametersJSON Schema
NameRequiredDescriptionDefault
source_datasetYes
target_fileNo

TDQS

B3.2/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided; description only states 'download', implying a read operation but fails to disclose any potential side effects, permissions, or limitations. Minimal behavioral context.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single sentence, no fluff, directly states purpose and format. Efficient and front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple tool with 2 params and no output schema, the description covers essential purpose but omits parameter details and usage context. Adequate but not comprehensive.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 0%, and description does not elaborate on parameters. 'source_dataset' and 'target_file' are self-explanatory but lack format, constraints, or examples, requiring agent inference.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Describes specific verb 'download', resource 'dataset from mainframe', and format 'binary', clearly distinguishing from siblings like download_text and download_pds_members.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No explicit guidance on when to use, when not, or alternatives. The format name 'binary' implies its use case, but lacks direct comparison to other download tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

download_job_spoolC

Download spool output for a specific job

ParametersJSON Schema
NameRequiredDescriptionDefault
jobidYes
target_fileNo

TDQS

C2.6/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries full burden. It does not disclose side effects, permission requirements, error handling, or what happens if jobid is invalid. Only a generic download is described.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise (one sentence), but it omits important details. It is not overly verbose, but could be more informative without being longer.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given no output schema and sparse parameter info, the description is incomplete. It fails to explain what spool output is, the format, or the purpose of target_file. Agent would lack key information.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so description must compensate. The description adds no meaning beyond the schema: jobid and target_file are uncommented. Target_file's role and default behavior (null) are unclear.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states the action 'Download' and resource 'spool output for a specific job', which is clear. However, it does not distinguish from sibling download tools like download_binary or download_text, which might also handle spool output.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance on when to use this tool versus alternatives. There is no mention of prerequisites, when to choose this tool over others, or exclusions.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

download_pds_membersC

Download members from a partitioned dataset (PDS)

ParametersJSON Schema
NameRequiredDescriptionDefault
datasetYes
target_dirNo
membersNo*
retr_modeNoascii
ftp_threadsNo
extensionNo
line_endingNo
preserve_trailing_spacesNo

TDQS

C2.8/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description bears full responsibility for disclosing behavioral traits. It only states the action without revealing side effects, authentication needs, or file handling behavior (e.g., overwrite rules). This leaves significant ambiguity.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single sentence, which is concise but under-specified for a tool with 8 parameters. It achieves brevity at the cost of completeness, earning an average score.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness1/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (8 parameters, no output schema, no annotations), the description is severely lacking. It does not explain return values, behavior, or how parameters affect the download. This is insufficient for an agent to use the tool correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The description adds no explanation for any of the 8 parameters. With 0% schema description coverage, the burden is higher; the description merely restates the tool's purpose, not the meaning or constraints of fields like 'members', 'retr_mode', or 'ftp_threads'.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description clearly specifies the action ('Download') and the resource ('members from a partitioned dataset (PDS)'). This distinguishes it from sibling tools like download_binary and download_text, which target different resource types.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives, no usage conditions, and no exclusions. Sibling tools exist for different download purposes, but the description does not hint at selection criteria.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

download_textC

Download a dataset from mainframe in text format

ParametersJSON Schema
NameRequiredDescriptionDefault
source_datasetYes
target_fileNo
line_endingNo
preserve_trailing_spacesNo

TDQS

C2.7/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description bears full responsibility for behavioral disclosure. It only mentions downloading a dataset in text format, failing to detail behavior such as connection handling, size limits, overwriting, encoding, or error conditions.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single sentence, which is concise, but it lacks necessary detail. It is not overly verbose, but brevity comes at the cost of informativeness.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has four parameters (one required) and no output schema or annotations, the description fails to cover return values, parameter dependencies, or behavior beyond a general statement. It is incomplete for effective use.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, yet the description adds no explanation for any of the four parameters (source_dataset, target_file, line_ending, preserve_trailing_spaces). The description only repeats the format, adding no value beyond what the schema provides.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb 'download', the resource 'dataset', the source 'mainframe', and the format 'text', distinguishing it from sibling tools like download_binary and download_job_spool.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No explicit guidance on when to use this tool versus alternatives like download_binary or download_pds_members. The description implies text format usage but omits context or prerequisites.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

get_connection_infoB

Get current connection information for the mainframe

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

B3.1/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so the description bears full burden. It only says 'Get current connection information' without disclosing side effects, restrictions, or any behavioral traits. The agent has no insight into read-only nature or dependencies.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single sentence, no wasted words. Perfectly concise.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema and no details on what 'connection information' includes. For a simple info tool, more specifics (e.g., 'including host, port, status') would improve completeness. Description is vague.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Tool has zero parameters, so baseline is 4. Schema description coverage is 100% trivially. Description adds no parameter info but none is needed.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description 'Get current connection information for the mainframe' clearly states the verb (get) and resource (connection information). It is distinct from sibling tools, which are all download/list operations, so no confusion.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance on when to use or when to avoid this tool. No mention of alternatives or context. The description only states what it does, not when to invoke it.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

get_job_infoC

Get job information and spool files for a specific job ID

ParametersJSON Schema
NameRequiredDescriptionDefault
jobidYes
jobmaskNo*

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It only says 'get job information and spool files' without explaining what information is included, whether spool files are returned as content or references, or any behavioral details like authentication or side effects. This is insufficient for a tool with no annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single sentence that gets to the point. It is efficient, though it could benefit from a bit more detail without becoming verbose. It earns its place by being direct.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple tool with 2 parameters and no output schema, the description is minimally adequate. However, it lacks clarity on what 'spool files' means (e.g., metadata or content) and does not explain the optional 'jobmask' parameter. Given the sibling 'download_job_spool', the description should clarify the relationship.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, meaning no parameter descriptions are provided. The description does not explain the meaning or format of 'jobid' or the optional 'jobmask' parameter, leaving the agent without guidance on how to fill in the arguments correctly.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states what the tool does: 'Get job information and spool files for a specific job ID'. It uses a specific verb and resource, and distinguishes from siblings like 'download_job_spool' (which downloads a spool file) and 'list_jes_jobs' (which lists jobs).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance is provided on when to use this tool versus alternatives. It does not specify when to prefer this over 'download_job_spool' for spool files or 'list_jes_jobs' for job listings. Usage context is only implied by the tool's purpose.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

list_catalogB

List datasets from mainframe catalog matching a pattern with pagination support

ParametersJSON Schema
NameRequiredDescriptionDefault
patternNoSYS1.*
limitNo
offsetNo

TDQS

B3.2/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries full burden. It mentions pagination support but does not disclose behavioral traits such as read-only nature, potential side effects, authentication requirements, or return value details beyond listing.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single sentence, front-loaded with verb and resource, no extraneous words. Efficiently conveys core purpose and key features.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool appears straightforward, but lacking output schema and behavioral details, the description leaves gaps about return format, pagination mechanics, and edge cases. It is minimally adequate for a simple list operation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, yet the description does not explain the parameters beyond hinting at 'pattern' and 'pagination'. The meanings of 'limit' and 'offset' are inferable but under-documented, and 'pattern' defaults are not clarified.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly identifies the action ('List'), the resource ('datasets from mainframe catalog'), and specific features (pattern matching, pagination support). It effectively distinguishes from sibling tools that handle downloads or job info.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No explicit guidance on when to use this tool versus alternatives, no prerequisites mentioned, and no exclusions stated. The description implies usage for listing with pattern and pagination but offers no comparative context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

list_jes_jobsC

List jobs from JES spool with pagination support

ParametersJSON Schema
NameRequiredDescriptionDefault
jobmaskNo
ownerNo
statusNoALL
limitNo
offsetNo

TDQS

C2.8/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so description must fully disclose behavioral traits, but it only hints at pagination without detailing filters, defaults, or authorization requirements.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single sentence is concise but too minimal; lacks structure and detailed information that could be included without verbosity.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness1/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 5 parameters, no output schema, and no annotations, the description is highly incomplete; agent lacks essential context to use the tool correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0% and description fails to explain any of the 5 parameters beyond mentioning pagination; does not compensate for missing parameter info.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description clearly states 'List jobs from JES spool', specifying verb and resource, and adds 'with pagination support', which distinguishes it from sibling tools like download_binary or get_job_info.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance on when to use this tool versus siblings; no mentions of prerequisites, filtering context, or when not to use.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

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

  1. 8 tool updatesv0.1.3
    • First observeddownload_binary
    • First observeddownload_job_spool
    • First observeddownload_pds_members
    • First observeddownload_text
    • First observedget_connection_info
    • First observedget_job_info
    • First observedlist_catalog
    • First observedlist_jes_jobs
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

B3.2/5.0

Scored across 8 tools

Disambiguation5/5

Each tool targets a distinct operation: downloading datasets in binary or text, downloading job spool, downloading PDS members, listing datasets or jobs, and retrieving connection or job info. There is no ambiguity between tools.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern using snake_case (e.g., download_binary, list_catalog). The verbs are imperative and clearly describe the action, and the nouns specify the resource.

Tool Count5/5

With 8 tools, the server covers the main operations for interacting with a mainframe via FTP (listing catalogs and jobs, downloading datasets in two formats, job spool, and PDS members, plus connection info). The count is well-scoped and not excessive.

Completeness2/5

The tool set is heavily read-oriented, missing upload operations (e.g., upload_binary, upload_text), the ability to list PDS members, and job management tools like submit or delete. These gaps significantly limit the server's usefulness for full FTP workflows.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • mainframe-mcp

    sagar25k
    A
    license
    Not graded
    quality
    D
    maintenance
    Provides safe, structured access to an IBM z/OS mainframe through natural language in Claude Code, enabling screen reading, dataset browsing, code editing, job submission, and regression testing.
    2
    MIT

  • ibmi-mcp-server

    nemu90
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables AI agents to interact with IBM i (AS/400) systems for RPA automation via 37 tools for 5250 navigation, CL commands, DB2 queries, and system management.
    MIT

  • ftpmcp

    vielhuber
    A
    license
    Not graded
    quality
    B
    maintenance
    An MCP server for FTP and FTPS file operations, supporting multiple named hosts with configurable roots, read-only mode, and guarded deletion. It enables remote file management (list, upload, download, create, rename, delete) through natural language while keeping credentials secure.
    13 npm
    MIT

  • mvsmf-mcp-bridge

    locusf
    A
    license
    Not graded
    quality
    B
    maintenance
    Enables AI assistants to interact with a classic MVS 3.8j system running under Hercules through mvsMF's z/OSMF-compatible REST API. Supports dataset listing/reading, member access, job submission, and job status/file retrieval via MCP tools.
    MIT