Skip to main content
Glama
osherai

Bullhorn CRM MCP Server

by osherai
OverviewSchemaRelated ServersScoreDiscussions
Python
Remote

Bullhorn CRM MCP Server

A Python Model Context Protocol (MCP) server that enables AI assistants to query your Bullhorn CRM data using natural language.

Works with: Claude Desktop, Claude Code, Cursor, Windsurf, Cline, Continue, Zed, and any MCP-compatible client.

This is an open-source alternative to paid connectors - it connects directly to Bullhorn's REST API with no additional subscriptions required.

Brought to you by Osher Digital - Specialist AI consultants helping businesses harness the power of artificial intelligence.

Features

  • Direct API Access - Connects to Bullhorn's REST API using OAuth 2.0

  • Natural Language Queries - Ask questions like "Show me the last 10 open jobs"

  • 6 Powerful Tools:

    • list_jobs - List and filter job orders

    • list_candidates - List and filter candidates

    • get_job - Get detailed job information by ID

    • get_candidate - Get detailed candidate information by ID

    • search_entities - Search any Bullhorn entity with Lucene queries

    • query_entities - Query entities with SQL-like WHERE syntax

  • Automatic Token Management - Handles OAuth token refresh automatically

  • Read-Only Access - Safe to use, no risk of modifying your CRM data

Related MCP server: Zoho CRM MCP Server

Prerequisites

  • Python 3.10+

  • uv (recommended) or pip

  • Bullhorn CRM account with API access

  • Bullhorn API credentials (Client ID, Client Secret, Username, Password)

Getting Your Bullhorn API Credentials

You'll need four credentials from Bullhorn:

  1. Client ID and Client Secret - OAuth application credentials

  2. API Username and API Password - Service account for API access

To obtain these:

  1. Contact your Bullhorn administrator or account manager

  2. Request API access for your account

  3. They will provide you with OAuth client credentials

  4. Create or use an existing service account for API authentication

Note: Your API username/password may be different from your regular Bullhorn login credentials.

Installation

1. Clone the Repository

git clone https://github.com/osherai/bullhorn-mcp-python.git
cd bullhorn-mcp-python

2. Install Dependencies

Using uv (recommended):

uv venv && uv pip install -e .

Or using pip:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -e .

3. Configure Credentials

Copy the example environment file and add your credentials:

cp .env.example .env

Edit .env with your Bullhorn API credentials:

BULLHORN_CLIENT_ID=your_client_id
BULLHORN_CLIENT_SECRET=your_client_secret
BULLHORN_USERNAME=your_api_username
BULLHORN_PASSWORD=your_api_password

4. Test the Connection

.venv/bin/python -c "
from bullhorn_mcp.config import BullhornConfig
from bullhorn_mcp.auth import BullhornAuth
from bullhorn_mcp.client import BullhornClient

config = BullhornConfig.from_env()
auth = BullhornAuth(config)
client = BullhornClient(auth)

jobs = client.search('JobOrder', 'isDeleted:0', count=3)
print(f'Successfully connected! Found {len(jobs)} jobs.')
"

Client Configuration

This MCP server works with any MCP-compatible client. Below are setup instructions for popular clients.

Note: Replace /path/to/bullhorn-mcp-python with your actual installation path in all examples below.

Claude Desktop

Add to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "bullhorn": {
      "command": "/path/to/bullhorn-mcp-python/.venv/bin/python",
      "args": ["-m", "bullhorn_mcp.server"],
      "cwd": "/path/to/bullhorn-mcp-python"
    }
  }
}

Restart Claude Desktop (fully quit and reopen) for changes to take effect.

Claude Code (CLI)

Add the server using the Claude Code CLI:

claude mcp add bullhorn \
  -e BULLHORN_CLIENT_ID=your_client_id \
  -e BULLHORN_CLIENT_SECRET=your_client_secret \
  -e BULLHORN_USERNAME=your_username \
  -e BULLHORN_PASSWORD=your_password \
  -- /path/to/bullhorn-mcp-python/.venv/bin/python -m bullhorn_mcp.server

Or add to your ~/.claude/settings.json:

{
  "mcpServers": {
    "bullhorn": {
      "command": "/path/to/bullhorn-mcp-python/.venv/bin/python",
      "args": ["-m", "bullhorn_mcp.server"],
      "cwd": "/path/to/bullhorn-mcp-python"
    }
  }
}

Cursor

Add to your Cursor MCP configuration:

macOS: ~/.cursor/mcp.json Windows: %USERPROFILE%\.cursor\mcp.json

{
  "mcpServers": {
    "bullhorn": {
      "command": "/path/to/bullhorn-mcp-python/.venv/bin/python",
      "args": ["-m", "bullhorn_mcp.server"],
      "cwd": "/path/to/bullhorn-mcp-python"
    }
  }
}

Restart Cursor for changes to take effect.

Windsurf (Codeium)

Add to your Windsurf MCP configuration:

macOS: ~/.codeium/windsurf/mcp_config.json Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json

{
  "mcpServers": {
    "bullhorn": {
      "command": "/path/to/bullhorn-mcp-python/.venv/bin/python",
      "args": ["-m", "bullhorn_mcp.server"],
      "cwd": "/path/to/bullhorn-mcp-python"
    }
  }
}

Restart Windsurf for changes to take effect.

VS Code with Cline Extension

Add to your Cline MCP settings:

macOS: ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json Windows: %APPDATA%\Code\User\globalStorage\saoudrizwan.claude-dev\settings\cline_mcp_settings.json

{
  "mcpServers": {
    "bullhorn": {
      "command": "/path/to/bullhorn-mcp-python/.venv/bin/python",
      "args": ["-m", "bullhorn_mcp.server"],
      "cwd": "/path/to/bullhorn-mcp-python"
    }
  }
}

VS Code with Continue Extension

Add to your Continue configuration at ~/.continue/config.json:

{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "/path/to/bullhorn-mcp-python/.venv/bin/python",
          "args": ["-m", "bullhorn_mcp.server"],
          "cwd": "/path/to/bullhorn-mcp-python"
        }
      }
    ]
  }
}

Zed Editor

Add to your Zed settings at ~/.config/zed/settings.json:

{
  "context_servers": {
    "bullhorn": {
      "command": {
        "path": "/path/to/bullhorn-mcp-python/.venv/bin/python",
        "args": ["-m", "bullhorn_mcp.server"]
      },
      "settings": {}
    }
  }
}

Example Queries

Once configured, you can ask natural language questions about your Bullhorn data:

  • "List the last 10 open jobs"

  • "Find candidates with Python experience"

  • "Show me details for job #12345"

  • "Search for active candidates added this month"

  • "What placements were made last week?"

Tools Reference

list_jobs

List and filter job orders from Bullhorn CRM.

Parameters:

Parameter

Type

Required

Description

query

string

No

Lucene search query

status

string

No

Filter by job status

limit

integer

No

Max results (default: 20, max: 500)

fields

string

No

Comma-separated fields to return

Examples:

list_jobs()                                    # Recent jobs
list_jobs(query="isOpen:1")                   # Open jobs only
list_jobs(query="title:Engineer", limit=10)  # Engineer jobs
list_jobs(status="Accepting Candidates")      # By status

list_candidates

List and filter candidates from Bullhorn CRM.

Parameters:

Parameter

Type

Required

Description

query

string

No

Lucene search query

status

string

No

Filter by candidate status

limit

integer

No

Max results (default: 20, max: 500)

fields

string

No

Comma-separated fields to return

Examples:

list_candidates()                              # Recent candidates
list_candidates(query="skillSet:Python")      # Python developers
list_candidates(status="Active", limit=50)    # Active candidates

get_job

Get detailed information for a specific job order.

Parameters:

Parameter

Type

Required

Description

job_id

integer

Yes

The JobOrder ID

fields

string

No

Comma-separated fields to return

get_candidate

Get detailed information for a specific candidate.

Parameters:

Parameter

Type

Required

Description

candidate_id

integer

Yes

The Candidate ID

fields

string

No

Comma-separated fields to return

search_entities

Search any Bullhorn entity type using Lucene query syntax.

Parameters:

Parameter

Type

Required

Description

entity

string

Yes

Entity type (JobOrder, Candidate, Placement, etc.)

query

string

Yes

Lucene search query

limit

integer

No

Max results (default: 20, max: 500)

fields

string

No

Comma-separated fields to return

Supported Entities:

  • JobOrder - Job postings

  • Candidate - Candidates/applicants

  • Placement - Job placements

  • ClientCorporation - Client companies

  • ClientContact - Client contacts

  • JobSubmission - Candidate submissions to jobs

  • Appointment - Scheduled appointments

  • Note - Notes and comments

  • And many more...

query_entities

Query Bullhorn entities using SQL-like WHERE syntax.

Parameters:

Parameter

Type

Required

Description

entity

string

Yes

Entity type

where

string

Yes

WHERE clause

limit

integer

No

Max results (default: 20, max: 500)

fields

string

No

Comma-separated fields to return

order_by

string

No

Sort order (e.g., "-dateAdded")

Examples:

query_entities(entity="JobOrder", where="salary > 100000")
query_entities(entity="Candidate", where="status='Active'", order_by="-dateAdded")

Query Syntax

Lucene Search Syntax

Used by list_jobs, list_candidates, and search_entities:

title:Engineer                           # Field contains value
isOpen:1                                 # Boolean/numeric field
salary:[50000 TO 100000]                # Range query
firstName:"John"                         # Exact phrase
firstName:John AND lastName:Smith       # AND condition
status:Active OR status:Available       # OR condition
NOT status:Inactive                      # Negation
name:Acme*                              # Wildcard

SQL-like WHERE Syntax

Used by query_entities:

salary > 100000                          # Comparison
status = 'Active'                        # Equality (use single quotes)
dateAdded > '2024-01-01'                # Date comparison
id IN (1, 2, 3, 4, 5)                   # IN clause
firstName = 'John' AND salary > 50000   # AND condition

Note: The LIKE operator is not supported in Bullhorn's query endpoint.

Default Fields

When fields is not specified, the following fields are returned:

JobOrder: id, title, status, employmentType, dateAdded, startDate, salary, clientCorporation, owner, description, numOpenings, isOpen

Candidate: id, firstName, lastName, email, phone, status, dateAdded, occupation, skillSet, owner

Environment Variables

Variable

Required

Description

BULLHORN_CLIENT_ID

Yes

OAuth 2.0 Client ID

BULLHORN_CLIENT_SECRET

Yes

OAuth 2.0 Client Secret

BULLHORN_USERNAME

Yes

API Username

BULLHORN_PASSWORD

Yes

API Password

BULLHORN_AUTH_URL

No

Auth URL (default: https://auth.bullhornstaffing.com)

BULLHORN_LOGIN_URL

No

Login URL (default: https://rest.bullhornstaffing.com)

Project Structure

bullhorn-mcp-python/
├── pyproject.toml              # Project configuration and dependencies
├── .env.example                # Environment variables template
├── README.md                   # This file
├── LICENSE                     # MIT License
└── src/
    └── bullhorn_mcp/
        ├── __init__.py         # Package initialization
        ├── server.py           # MCP server with tool definitions
        ├── auth.py             # Bullhorn OAuth 2.0 authentication
        ├── client.py           # Bullhorn REST API client
        └── config.py           # Configuration management

Troubleshooting

"Missing required environment variables"

Ensure all required variables are set in your .env file or environment.

Authentication Errors

  1. Verify your credentials are correct

  2. Check that your API user has appropriate permissions

  3. Ensure your Bullhorn account has API access enabled

"Connection refused" or timeout errors

  1. Check your internet connection

  2. Verify the auth/login URLs are correct for your Bullhorn datacenter

  3. Some Bullhorn instances use regional URLs (e.g., rest9.bullhornstaffing.com)

MCP server not appearing in your client

  1. Ensure the config file path is correct for your client (see Client Configuration section)

  2. Verify the Python path in the config points to the .venv directory

  3. Fully quit and restart your client application

  4. Check your client's logs for error messages

  5. Test the server manually:

    cd /path/to/bullhorn-mcp-python
.venv/bin/python -m bullhorn_mcp.server

    The server should start without errors (it will wait for input on stdin)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

  1. Fork the repository

  2. Create your feature branch (git checkout -b feature/amazing-feature)

  3. Commit your changes (git commit -m 'Add some amazing feature')

  4. Push to the branch (git push origin feature/amazing-feature)

  5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

About Osher Digital

This project is maintained by Osher Digital, specialist AI consultants based in Australia. We help businesses integrate AI solutions to streamline operations and drive growth.

Need help with AI integration? Get in touch

Disclaimer

This is an unofficial, community-maintained project. It is not affiliated with, officially maintained, or endorsed by Bullhorn.

Available Tools

6 tools
get_candidateA

Get details for a specific candidate by ID.

Args: candidate_id: The Candidate ID fields: Comma-separated fields to return (default: all common fields)

Returns: JSON object with candidate details

ParametersJSON Schema
NameRequiredDescriptionDefault
candidate_idYes
fieldsNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.3/5.0
Behavior4/5

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

No annotations are provided, so the description carries full responsibility. It states it returns a JSON object with candidate details, implying a read operation. However, it does not mention error handling, permissions, or idempotency specifics, though the basic read behavior is clear.

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?

The description is concise and well-structured with 'Args' and 'Returns' sections. Every sentence is informative, with no redundant content.

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

Completeness4/5

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

Given the presence of an output schema (not shown) and two well-described parameters, the description is mostly complete for a basic get operation. However, it could mention the typical workflow (e.g., use 'list_candidates' to obtain IDs) and provide some error handling notes.

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

Parameters5/5

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

Schema description coverage is 0%, so the description must compensate. It explains 'candidate_id' as 'The Candidate ID' and 'fields' as 'Comma-separated fields to return (default: all common fields)', adding meaningful semantics beyond the schema's titles and types.

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 'Get details for a specific candidate by ID,' using a specific verb and resource. It distinguishes from siblings like 'list_candidates' and 'get_job' by targeting a single candidate via ID.

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

Usage Guidelines3/5

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

The description implies usage when you have a candidate ID but does not explicitly state when to use this tool versus alternatives like 'list_candidates' to find IDs. No exclusions or alternative recommendations are provided.

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

get_jobA

Get details for a specific job order by ID.

Args: job_id: The JobOrder ID fields: Comma-separated fields to return (default: all common fields)

Returns: JSON object with job details

ParametersJSON Schema
NameRequiredDescriptionDefault
job_idYes
fieldsNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.1/5.0
Behavior3/5

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

No annotations provided, so description carries the burden. It states the return type (JSON object) but does not mention error handling (e.g., job not found), authentication requirements, or side effects. For a simple read tool, it is minimally adequate.

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?

The description is very concise, using a docstring format with Args and Returns sections. It is front-loaded with the purpose and contains no redundant words.

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

Completeness4/5

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

Given the tool's simplicity (2 params, no pagination), the description covers input and output adequately. An output schema exists. It could mention behavior on missing job ID, but overall it is sufficiently complete.

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

Parameters5/5

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

Schema description coverage is 0%, but the description adds full semantics: job_id is the ID, and fields is a comma-separated string with default behavior explained. This compensates completely for the missing schema descriptions.

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 tool gets details for a specific job by ID, using a clear verb and resource. It distinguishes itself from siblings like list_jobs and get_candidate.

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

Usage Guidelines3/5

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

Usage is implied: use when you have a job ID and want details. But there is no explicit guidance on when not to use it or alternatives, such as using list_jobs for browsing.

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

list_candidatesA

List and filter candidates from Bullhorn CRM.

Args: query: Lucene search query (e.g., "lastName:Smith" or "skillSet:Python") status: Filter by candidate status limit: Maximum number of results (1-500, default 20) fields: Comma-separated fields to return

Returns: JSON array of candidates

Examples: - list_candidates() - Get recent candidates - list_candidates(query="skillSet:Python") - Find Python developers - list_candidates(query="lastName:Smith AND status:Active") - list_candidates(status="Active", limit=50)

ParametersJSON Schema
NameRequiredDescriptionDefault
queryNo
statusNo
limitNo
fieldsNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.3/5.0
Behavior3/5

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

No annotations are provided, so the description must fully disclose behavior. It states returns a JSON array and explains parameters, but lacks details on pagination, sorting, error handling, or response size limits beyond the implied limit param. The examples help but do not cover all behavioral traits.

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?

The description is well-structured with a summary, Args, Returns, and Examples sections. It is concise yet informative, using bullet points and code examples to aid comprehension without redundancy.

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

Completeness4/5

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

The description covers all parameters, return type, and usage patterns with examples. However, it omits details on error conditions, authentication requirements, or data freshness. Given moderate complexity and presence of an output schema, it is nearly complete.

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

Parameters5/5

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

Schema coverage is 0%, so the description must compensate. It defines each parameter with usage: query includes Lucene syntax examples, status is a filter, limit has range and default, fields describes comma-separated selection. This fully explains semantics beyond the schema.

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 tool's purpose: 'List and filter candidates from Bullhorn CRM.' It specifies the action (list/filter), the resource (candidates), and the source. This distinguishes it from sibling tools like get_candidate (single candidate) and list_jobs (different entity).

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

Usage Guidelines4/5

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

The description includes parameter explanations and four usage examples, giving clear context for typical use cases. However, it does not explicitly state when to use this tool vs. alternatives (e.g., 'use get_candidate for a single candidate'), leaving some ambiguity.

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

list_jobsA

List and filter job orders from Bullhorn CRM.

Args: query: Lucene search query (e.g., "title:Engineer AND isOpen:1") status: Filter by job status limit: Maximum number of results (1-500, default 20) fields: Comma-separated fields to return

Returns: JSON array of job orders

Examples: - list_jobs() - Get recent jobs - list_jobs(query="isOpen:1") - Get open jobs - list_jobs(query="title:Software AND employmentType:Direct Hire", limit=10) - list_jobs(status="Accepting Candidates")

ParametersJSON Schema
NameRequiredDescriptionDefault
queryNo
statusNo
limitNo
fieldsNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.5/5.0
Behavior4/5

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

No annotations provided, so description fully explains behavior: returns JSON array, supports Lucene query, defaults to 20 results. It doesn't mention pagination or permissions, but non-destructive listing is implied.

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?

Description is structured with Args, Returns, and Examples. It is concise enough for a tool with 4 parameters, though slightly verbose. Front-loaded with purpose, making it easy to scan.

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

Completeness5/5

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

Given output schema exists and no annotations, description is complete: covers all parameters, provides examples, mentions defaults and return format. No gaps for this complexity level.

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

Parameters5/5

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

Schema has 0% description coverage, but description compensates fully: explains query (Lucene), status (filter), limit (max 1-500, default 20), fields (comma-separated). Examples demonstrate usage, adding significant value.

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 it lists and filters job orders from Bullhorn CRM. It distinguishes from sibling tools like get_job (single job) and list_candidates (different entity), making its purpose specific.

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

Usage Guidelines4/5

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

Provides examples showing different usage patterns (no args, query, status), giving clear context. However, it does not explicitly state when not to use this tool or mention alternatives like query_entities.

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

query_entitiesA

Query Bullhorn entities using SQL-like WHERE syntax.

Args: entity: Entity type (JobOrder, Candidate, etc.) where: WHERE clause (e.g., "salary > 100000 AND status='Active'") limit: Maximum number of results (1-500, default 20) fields: Comma-separated fields to return order_by: Sort order (e.g., "-dateAdded" for newest first)

Returns: JSON array of matching entities

Examples: - query_entities(entity="JobOrder", where="salary > 100000") - query_entities(entity="Candidate", where="status='Active'", order_by="-dateAdded")

ParametersJSON Schema
NameRequiredDescriptionDefault
entityYes
whereYes
limitNo
fieldsNo
order_byNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4/5.0
Behavior3/5

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

With no annotations, the description carries full burden. It explains return format ('JSON array of matching entities') and parameter behaviors (limit range, default). However, it does not disclose error handling, authentication requirements, or behavior for invalid entity types, leaving gaps.

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 well-structured with sections (Args, Returns, Examples) and is mostly concise. Each sentence adds value. Minor redundancy (e.g., 'JSON array' could be implied) but overall efficient.

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

Completeness4/5

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

Given the tool's complexity (5 params, output schema exists), the description covers parameter details, return type, and examples. It could mention valid entity types explicitly, but the output schema may cover that. Overall complete for a query tool.

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

Parameters5/5

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

Schema description coverage is 0%, so the description must compensate. It clearly explains each parameter: entity type with examples, WHERE clause syntax, limit range, fields as comma-separated, and order_by format. Examples further illustrate usage. This adds significant meaning.

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 tool's purpose: 'Query Bullhorn entities using SQL-like WHERE syntax.' It specifies the resource (Bullhorn entities) and the action (querying with WHERE syntax). Examples further clarify. It distinguishes from sibling tools like get_candidate or list_candidates by emphasizing the SQL-like query capability.

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

Usage Guidelines3/5

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

The description implies usage for complex filtering via WHERE syntax but does not explicitly contrast with siblings like search_entities or list_* tools. No when-not or alternative guidance is provided. Examples help, but explicit comparison is missing.

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

search_entitiesA

Search any Bullhorn entity type using Lucene query syntax.

Args: entity: Entity type (JobOrder, Candidate, Placement, ClientCorporation, ClientContact, etc.) query: Lucene search query limit: Maximum number of results (1-500, default 20) fields: Comma-separated fields to return

Returns: JSON array of matching entities

Examples: - search_entities(entity="Placement", query="status:Approved") - search_entities(entity="ClientCorporation", query="name:Acme*") - search_entities(entity="JobSubmission", query="jobOrder.id:12345")

ParametersJSON Schema
NameRequiredDescriptionDefault
entityYes
queryYes
limitNo
fieldsNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.3/5.0
Behavior4/5

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

Despite no annotations, the description is transparent: it explains the search functionality, returns a JSON array, and uses Lucene syntax. It does not explicitly state read-only behavior, but the examples imply no side effects. The description adequately discloses the core behavior.

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?

The description is concise and well-structured: a one-line summary, then Args, Returns, and Examples. Every sentence is purposeful, no redundancy. The format is front-loaded and easy to parse.

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

Completeness4/5

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

The description covers the essential aspects: entity types, query syntax, limit, fields, and return format. With an output schema present, the return description is sufficient. Minor omissions (e.g., pagination) are acceptable for a search tool.

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

Parameters5/5

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

With 0% schema coverage, the description fully compensates by detailing each parameter: entity types, Lucene query, limit range (1-500, default 20), and fields as comma-separated. Examples illustrate valid values, adding substantial meaning beyond the schema.

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 starts with 'Search any Bullhorn entity type using Lucene query syntax,' which clearly specifies the action (search) and the resource (any entity type). It distinguishes from sibling tools like get_candidate and list_candidates, which retrieve single entities or lists without search syntax.

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

Usage Guidelines3/5

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

The description implies usage when a Lucene query is needed, but it does not explicitly state when to use this tool versus alternatives like query_entities. No exclusions or when-not-to-use guidance is provided, leaving the agent to infer context from examples.

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. Dates show when Glama detected each change.

  1. 6 tool updatesv0.1.0
    • First observedget_candidate
    • First observedget_job
    • First observedlist_candidates
    • First observedlist_jobs
    • First observedquery_entities
    • First observedsearch_entities
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4/5.0
Disambiguation3/5

Tool purposes are mostly distinct, but list_candidates/list_jobs overlap with search_entities since search_entities can also query candidates/jobs with Lucene syntax, creating potential confusion for an agent.

Naming Consistency4/5

Names follow a verb_noun pattern (get_, list_, query_, search_), but verbs are not uniform (get vs list, query vs search). This is a minor inconsistency.

Tool Count5/5

Six tools is appropriate for a read-only CRM data access layer. Each tool has a clear purpose without unnecessary bloat.

Completeness2/5

The server covers only read operations (get, list, query, search) and provides no create, update, or delete tools. For a CRM server, this is a significant gap that will limit agent workflows.

Maintenance

ActivityInactive
ResponsivenessUnresponsive

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • HaloPSA MCP Server

    Switchboard666
    F
    license
    A
    quality
    Not graded
    maintenance
    Enables AI assistants to interact with HaloPSA data through secure OAuth2 authentication. Supports SQL queries against the HaloPSA database, API endpoint exploration, and direct API calls for comprehensive PSA data analysis and management.
    10
    7
    -

  • Zoho CRM MCP Server

    whiteside-daniel
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables read-only interaction with Zoho CRM data through natural language queries, allowing users to search records, list modules, retrieve field information, and count records using secure OAuth authentication.
    2
    -

  • MCP Salesforce Lite

    luvl
    A
    license
    A
    quality
    D
    maintenance
    Enables AI assistants to securely interact with Salesforce CRM data through SOQL queries, CRUD operations, and metadata exploration. Supports connecting to Salesforce objects like Accounts, Contacts, and Opportunities via OAuth 2.0 authentication.
    8
    2
    MIT

  • Salesforce MCP Server

    tweiss777
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables AI assistants to interact with Salesforce CRM by executing SOQL queries and performing CRUD operations on records such as Leads. It supports secure OAuth 2.0 authentication and provides management for both standard and custom Salesforce fields.
    -

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/osherai/bullhorn-mcp-python'

If you have feedback or need assistance with the MCP directory API, please join our Discord server