Skip to main content
Glama
Freshservice MCP Server

Freshservice MCP Server

by matthewlboyd

Freshservice MCP Server

A Model Context Protocol (MCP) server that connects Claude, Codex, and other MCP clients to your Freshservice ITSM instance.

Supported modules

Module

Operations

Tickets

Create, view, list, update, delete, filter, get field definitions

Conversations

List, reply to ticket, add notes (public/private), delete notes

Changes

Create, view, list, update, close, delete, filter, tasks, notes

Assets

Create, view, list, update, delete, search, filter, get types

Problems

Create, view, list

Agents

List, view, filter

Requesters

Create, list, view, filter

Agent Groups

List, view

Requester Groups

List, view

Products

List, view

Workspaces

List

Canned Responses

List folders, list responses, view

Solution Categories

Create, list, view

Solution Folders

Create, list, view

Solution Articles

Create, list, view, search

Departments

List, view

Locations

List

Software

List, view

Vendors

List

Service Catalog

List items, view item

Announcements

List

Contracts

List, view

Purchase Orders

List

Related MCP server: Freshservice MCP Server

Prerequisites

  • Python 3.10+

  • uv — fast Python package manager

  • Claude Desktop, Claude Code, or Codex app/CLI/IDE (see Setup below)

  • A Freshservice account with an API key

Install uv

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

# Or with Homebrew
brew install uv

Get your Freshservice API key

  1. Log in to your Freshservice portal

  2. Click your profile picture → Profile Settings

  3. Your API key is shown on the right side under Change Password

Setup

1. Clone the repository

git clone https://github.com/matthewlboyd/freshservice-mcp.git
cd freshservice-mcp

2a. Configure Claude Desktop

Option A — macOS install script (recommended):

./install.sh

This handles everything automatically: prompts for your repo path, API key, and domain, writes the config, tests the connection, and restarts Claude Desktop.

Option B — Manual setup (all platforms):

Open the Claude Desktop config file in any text editor:

OS

Path

macOS

~/Library/Application Support/Claude/claude_desktop_config.json

Windows

%APPDATA%\Claude\claude_desktop_config.json

Linux

~/.config/Claude/claude_desktop_config.json

Paste the following into the file, replacing the placeholder values. If the file doesn't exist yet, create it with just this content:

{
  "mcpServers": {
    "freshservice": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/full/path/to/freshservice-mcp",
        "freshservice-mcp"
      ],
      "env": {
        "FRESHSERVICE_APIKEY": "your_api_key_here",
        "FRESHSERVICE_DOMAIN": "yourcompany.freshservice.com"
      }
    }
  }
}

FRESHSERVICE_DOMAIN should be just the hostname — no https:// prefix.

If you're behind a corporate proxy with SSL inspection, add SSL_CERT_FILE to the env block pointing at your CA bundle:

"env": {
  "FRESHSERVICE_APIKEY": "your_api_key_here",
  "FRESHSERVICE_DOMAIN": "yourcompany.freshservice.com",
  "SSL_CERT_FILE": "/path/to/your/ca-bundle.crt"
}

Already have other MCP servers? Add just the "freshservice": { ... } block inside your existing "mcpServers" object instead of replacing the whole file.

Restart Claude Desktop — quit and reopen it. The first launch will automatically install dependencies (takes ~30 seconds). You should see a hammer icon (🔨) in the chat input, indicating MCP tools are available.

2b. Configure Claude Code

Recommended: project-local scope

claude mcp add freshservice \
  --scope local \
  --env FRESHSERVICE_APIKEY=your_api_key_here \
  --env FRESHSERVICE_DOMAIN=yourcompany.freshservice.com \
  -- uv run --directory /full/path/to/freshservice-mcp freshservice-mcp

--scope local stores the config in .claude/settings.local.json, which is gitignored by default — your API key will not be committed to version control.

Alternative: user scope (available in all your projects)

claude mcp add freshservice \
  --scope user \
  --env FRESHSERVICE_APIKEY=your_api_key_here \
  --env FRESHSERVICE_DOMAIN=yourcompany.freshservice.com \
  -- uv run --directory /full/path/to/freshservice-mcp freshservice-mcp

Security note: Never use --scope project with credentials hardcoded in the command — the project scope writes to .mcp.json, which is typically committed to version control. Use local or user scope, or pass credentials via environment variables already set in your shell.

If you're behind a corporate proxy with SSL inspection, add --env SSL_CERT_FILE=/path/to/your/ca-bundle.crt to the command above.

Verify the server is registered:

claude mcp list

You should see freshservice in the output. No restart is needed — Claude Code picks up MCP servers at the start of each session.

2c. Configure Codex app, CLI, or IDE

Codex app, Codex CLI, and the Codex IDE extension share MCP configuration through config.toml. The server uses Codex-supported stdio transport and can run through the Codex-specific entrypoint alias freshservice-mcp-codex.

Option A — install script:

./install-codex.sh

This prompts for your repo path, Freshservice API key, domain, and optional CA bundle, then writes a [mcp_servers.freshservice] block to ~/.codex/config.toml (or $CODEX_HOME/config.toml if CODEX_HOME is set).

Option B — Codex CLI:

codex mcp add freshservice \
  --env FRESHSERVICE_APIKEY=your_api_key_here \
  --env FRESHSERVICE_DOMAIN=yourcompany.freshservice.com \
  -- uv run --directory /full/path/to/freshservice-mcp freshservice-mcp-codex

If you're behind a corporate proxy with SSL inspection, add --env SSL_CERT_FILE=/path/to/ca-bundle.crt.

Option C — manual config.toml:

Open Codex Settings → Integrations & MCP → Open config.toml, or edit ~/.codex/config.toml directly. Add this block, replacing the placeholders:

[mcp_servers.freshservice]
enabled = true
command = "uv"
args = [
  "run",
  "--directory",
  "/full/path/to/freshservice-mcp",
  "freshservice-mcp-codex",
]
startup_timeout_sec = 30
tool_timeout_sec = 120

[mcp_servers.freshservice.env]
FRESHSERVICE_APIKEY = "your_api_key_here"
FRESHSERVICE_DOMAIN = "yourcompany.freshservice.com"

FRESHSERVICE_DOMAIN should be just the hostname — no https:// prefix.

If you're behind a corporate proxy with SSL inspection, add this to the env block:

SSL_CERT_FILE = "/path/to/your/ca-bundle.crt"

Security note: Values under [mcp_servers.freshservice.env] are stored in plaintext in config.toml. If you already provide these variables to the Codex process another way, you can use env_vars = ["FRESHSERVICE_APIKEY", "FRESHSERVICE_DOMAIN"] instead of the env table.

Restart the Codex app, or start a new Codex CLI/IDE session. In Codex, use /mcp to confirm the freshservice server is loaded.

Example prompts

Tickets

  • "Show me all open urgent tickets"

  • "Create a ticket for a VPN connectivity issue — assign to the Network team, high priority"

  • "Update ticket #4521 to resolved"

  • "Find all tickets created this week by john@example.com"

Changes

  • "Create a change request for server maintenance this Saturday, 10pm–2am"

  • "Show all changes awaiting approval"

  • "Close change #89 — migration completed successfully"

Assets

  • "List all laptops in the asset inventory"

  • "Search for assets tagged 'datacenter'"

  • "Show all unassigned hardware assets"

Knowledge Base

  • "Search the knowledge base for VPN setup instructions"

  • "Create a new solution article on how to reset your password"

  • "List all articles in the Onboarding folder"

General

  • "How many open tickets are assigned to the Help Desk group?"

  • "List all active agents and their departments"

  • "Show me all software registered in the system"

Running the tests

The test suite covers all 62 tools and the API client (117 tests total). No Freshservice credentials are needed — all tests use mocked HTTP calls.

Install dev dependencies:

uv pip install -e ".[dev]"

Run all tests:

uv run pytest tests/ -v

Run a specific module's tests:

uv run pytest tests/test_tickets.py -v

Test files:

File

What it covers

test_client.py

HTTP client — auth, domain normalisation, error handling

test_tickets.py

Ticket tools

test_conversations.py

Reply, notes, delete note

test_changes.py

Change request tools

test_assets.py

Asset tools

test_people.py

Agents, requesters, groups

test_knowledge_base.py

Solution categories, folders, articles

test_misc.py

Problems, products, workspaces, canned responses, departments, locations, software, vendors, service catalog, announcements, contracts, purchase orders

Troubleshooting

Tools don't appear in Claude Desktop Make sure the path in --directory is the absolute path to the repo and that you've fully quit and restarted Claude (not just closed the window).

Tools don't appear in Claude Code Run claude mcp list to confirm the server is registered. If it is, start a new claude session — MCP servers are loaded at session start. If it's missing, re-run the claude mcp add command.

Tools don't appear in Codex Check ~/.codex/config.toml (or $CODEX_HOME/config.toml) for the [mcp_servers.freshservice] block, then restart the Codex app or start a new Codex CLI/IDE session. In Codex, run /mcp to inspect loaded MCP servers. If first launch times out while uv installs dependencies, keep startup_timeout_sec = 30 or increase it.

Authentication error (401) Your API key is incorrect. Double-check it in Freshservice → Profile Settings. Copy it using the copy icon rather than selecting the text manually.

403 Forbidden The API key doesn't have permission for that operation, or the domain is wrong. Confirm FRESHSERVICE_DOMAIN is just the hostname (e.g. yourcompany.freshservice.com), with no https:// prefix.

SSL certificate errors If you're behind a corporate proxy with SSL inspection, requests will fail with a certificate verification error. Set the SSL_CERT_FILE environment variable to the path of your organisation's CA bundle. In Claude Desktop, add it to the "env" block in claude_desktop_config.json. In Claude Code, pass --env SSL_CERT_FILE=/path/to/ca-bundle.crt when running claude mcp add. In Codex, add SSL_CERT_FILE = "/path/to/ca-bundle.crt" under [mcp_servers.freshservice.env].

Rate limiting (429) Freshservice enforces per-minute API rate limits based on your plan. Wait a moment and try again.

Filter queries returning errors Filter strings must use double-quoted field values. Example: "status:2 AND priority:4" not status:2 AND priority:4.

Checking logs If something isn't working, check the MCP server logs:

  • macOS: ~/Library/Logs/Claude/mcp-server-freshservice.log

  • Windows: %APPDATA%\Claude\logs\mcp-server-freshservice.log

License

MIT

Available Tools

70 tools
add_ticket_noteA

Add a note to a ticket.

Args: ticket_id: The ticket ID. body: Note body (HTML supported). private: True for private (agent-only) note, False for public.

Inputs: body (required), ticket_id (required), private. Returns: result

close_changeB

Close a change with a result explanation.

Args: change_id: The change ID. change_result_explanation: Explanation of the change result.

Inputs: change_id (required), change_result_explanation (required). Returns: result

create_assetA

Create a new asset.

Args: name: Asset name. asset_type_id: Asset type ID (from get_asset_types). description: Asset description. impact: 1=Low, 2=Medium, 3=High. user_id: User assigned to the asset. department_id: Department ID. location_id: Location ID. agent_id: Agent managing the asset. asset_tag: Custom asset tag. custom_fields: Custom field key-value pairs.

Inputs: name (required), asset_type_id (required), impact, user_id, +6 more. Returns: result

create_changeA

Create a new change request.

Args: requester_id: ID of the requester. subject: Change subject. description: Change description (HTML supported). priority: 1=Low, 2=Medium, 3=High, 4=Urgent. impact: 1=Low, 2=Medium, 3=High. status: 1=Open, 2=Planning, 3=Awaiting Approval, 4=Pending Release, 5=Pending Review, 6=Closed. risk: 1=Low, 2=Medium, 3=High, 4=Very High. change_type: 1=Minor, 2=Standard, 3=Major, 4=Emergency. planned_start_date: ISO 8601 datetime string. planned_end_date: ISO 8601 datetime string. group_id: Agent group ID. workspace_id: Workspace ID. custom_fields: Custom field key-value pairs.

Inputs: subject (required), description (required), requester_id (required), risk, +9 more. Returns: result

create_change_noteA

Add a note to a change request.

Args: change_id: The change ID. body: Note body (HTML supported). private: True for agent-only note, False for public.

Inputs: body (required), change_id (required), private. Returns: result

create_problemB

Create a new problem.

Args: subject: Problem subject. description: Problem description (HTML). requester_id: Requester ID. priority: 1=Low, 2=Medium, 3=High, 4=Urgent. impact: 1=Low, 2=Medium, 3=High. status: 1=Open, 2=Change Requested, 3=Closed. group_id: Agent group ID. custom_fields: Custom fields.

Inputs: subject (required), description (required), requester_id (required), impact, +4 more. Returns: result

create_requesterB

Create a new requester / contact.

Args: first_name: Requester first name. email: Requester email address. last_name: Requester last name. phone: Phone number. department_ids: List of department IDs. custom_fields: Custom fields.

Inputs: first_name (required), email, phone, last_name, +2 more. Returns: result

create_solution_articleA

Create a new solution article.

Args: folder_id: Parent folder ID. title: Article title. description: Article body (HTML supported). status: 1=Draft, 2=Published. article_type: 1=Permanent, 2=Workaround. tags: Optional list of tags.

Inputs: title (required), folder_id (required), description (required), tags, +2 more. Returns: result

create_solution_categoryC

Create a new solution category.

Args: name: Category name. description: Category description.

Inputs: name (required), description. Returns: result

create_solution_folderA

Create a new solution folder.

Args: category_id: Parent category ID. name: Folder name. description: Folder description. visibility: 1=All, 2=Logged-in users, 3=Agents only.

Inputs: name (required), category_id (required), visibility, description. Returns: result

create_ticketA

Create a new Freshservice ticket.

Args: subject: Ticket subject line. description: Ticket description (HTML supported). email: Requester email. Creates a new contact if not found. priority: 1=Low, 2=Medium, 3=High, 4=Urgent. status: 2=Open, 3=Pending, 4=Resolved, 5=Closed. source: 1=Email, 2=Portal, 3=Phone, 4=Chat, 9=Walkup, 10=Slack. type: Ticket type (default "Incident"). group_id: ID of assigned agent group. responder_id: ID of assigned agent. department_id: Department ID. category: Ticket category. sub_category: Ticket sub-category. tags: List of tags. cc_emails: CC email addresses. custom_fields: Custom field key-value pairs. workspace_id: Workspace ID (for multi-workspace accounts).

Inputs: email (required), subject (required), description (required), tags, +12 more. Returns: result

delete_assetA

Delete an asset (soft delete / trash).

Args: display_id: The asset display ID.

Inputs: display_id (required). Returns: result

delete_changeB

Delete a change request.

Args: change_id: The change ID.

Inputs: change_id (required). Returns: result

delete_ticketA

Trash a ticket (soft delete).

Args: ticket_id: The ticket ID to delete.

Inputs: ticket_id (required). Returns: result

delete_ticket_noteB

Delete a note from a ticket.

Args: ticket_id: The ticket ID. conversation_id: The conversation (note) ID to delete.

Inputs: ticket_id (required), conversation_id (required). Returns: result

filter_agentsC

Filter agents by various criteria.

Args: query: Filter query (e.g. email, name, department, active status). page: Page number. per_page: Results per page.

Inputs: query (required), page, per_page. Returns: result

filter_assetsB

Filter assets using Freshservice query language.

Args: query: Filter query. E.g. "asset_type_id:1", "department_id:5". page: Page number. per_page: Results per page.

Inputs: query (required), page, per_page. Returns: result

filter_changesA

Filter changes with advanced queries.

Query MUST be wrapped in double quotes. Examples: "status:3" — awaiting approval "approval_status:1" — approved "approval_status:1 AND status:<6" — approved but not closed

Args: query: Freshservice filter query string. page: Page number. per_page: Results per page.

Inputs: query (required), page, per_page. Returns: result

filter_requestersC

Filter requesters. Query examples: email, name, department, etc.

Args: query: Filter query string. page: Page number. per_page: Results per page.

Inputs: query (required), page, per_page. Returns: result

filter_ticketsA

Filter tickets using Freshservice query language.

The query MUST be wrapped in double quotes. Examples: "status:2" — open tickets "priority:4" — urgent tickets "agent_id:1234 AND status:2" "group_id:5 AND priority:>2" "created_at:>'2024-01-01'"

Args: query: Freshservice filter query string (must be in double quotes). page: Page number. per_page: Results per page (max 100).

Inputs: query (required), page, per_page. Returns: result

get_agent_by_idB

Retrieve a single agent by ID.

Args: agent_id: The agent ID.

Inputs: agent_id (required). Returns: result

get_agent_group_by_idB

Get details of a specific agent group.

Args: group_id: The group ID.

Inputs: group_id (required). Returns: result

get_agent_groupsB

List all agent groups.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_agentsA

List all agents with pagination.

Args: page: Page number. per_page: Results per page (max 100).

Inputs: page, per_page. Returns: result

get_announcementsA

List all announcements.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_asset_by_idA

Retrieve a single asset by its display ID.

Args: display_id: The asset display ID.

Inputs: display_id (required). Returns: result

get_assetsA

List all assets with pagination.

Args: page: Page number. per_page: Results per page (max 100).

Inputs: page, per_page. Returns: result

get_asset_typesA

List all asset types.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_canned_responseC

Get a specific canned response by ID.

Args: response_id: The canned response ID.

Inputs: response_id (required). Returns: result

get_canned_response_foldersA

List all canned response folders.

Inputs: no named parameters. Returns: result

get_canned_responses_in_folderA

List all canned responses within a folder.

Args: folder_id: The canned response folder ID.

Inputs: folder_id (required). Returns: result

get_change_by_idB

Retrieve a single change by ID.

Args: change_id: The change ID.

Inputs: change_id (required). Returns: result

get_changesA

List all changes with optional filtering.

Args: page: Page number. per_page: Results per page (max 100). query: Optional filter query, MUST be in double quotes. E.g. "status:3", "approval_status:1 AND status:<6".

Inputs: page, query, per_page. Returns: result

get_change_tasksC

Get all tasks associated with a change.

Args: change_id: The change ID.

Inputs: change_id (required). Returns: result

get_contract_by_idC

View a specific contract.

Args: contract_id: The contract ID.

Inputs: contract_id (required). Returns: result

get_contractsB

List all contracts.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_department_by_idB

View a specific department.

Args: department_id: The department ID.

Inputs: department_id (required). Returns: result

get_departmentsB

List all departments.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_locationsC

List all locations.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_problem_by_idB

Retrieve a single problem by ID.

Args: problem_id: The problem ID.

Inputs: problem_id (required). Returns: result

get_problemsA

List all problems with pagination.

Args: page: Page number. per_page: Results per page (max 100).

Inputs: page, per_page. Returns: result

get_product_by_idB

Get details of a specific product.

Args: product_id: The product ID.

Inputs: product_id (required). Returns: result

get_productsB

List all products.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_purchase_ordersB

List all purchase orders.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_requester_by_idB

Retrieve a single requester by ID.

Args: requester_id: The requester ID.

Inputs: requester_id (required). Returns: result

get_requester_group_by_idC

Get details of a specific requester group.

Args: group_id: The group ID.

Inputs: group_id (required). Returns: result

get_requester_groupsB

List all requester groups.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_requestersA

List all requesters with pagination.

Args: page: Page number. per_page: Results per page (max 100).

Inputs: page, per_page. Returns: result

get_service_catalog_itemB

View a specific service catalog item.

Args: item_id: The service catalog item ID.

Inputs: item_id (required). Returns: result

get_service_catalog_itemsB

List all service catalog items.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_softwareB

List all software entries.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_software_by_idC

View a specific software entry.

Args: software_id: The software ID.

Inputs: software_id (required). Returns: result

get_solution_article_by_idC

View a specific solution article.

Args: article_id: The article ID.

Inputs: article_id (required). Returns: result

get_solution_articlesA

List all solution articles in a folder.

Args: folder_id: The parent folder ID. page: Page number. per_page: Results per page.

Inputs: folder_id (required), page, per_page. Returns: result

get_solution_categoriesC

List all solution (knowledge base) categories.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_solution_category_by_idA

View a specific solution category.

Args: category_id: The category ID.

Inputs: category_id (required). Returns: result

get_solution_folder_by_idB

View a specific solution folder.

Args: folder_id: The folder ID.

Inputs: folder_id (required). Returns: result

get_solution_foldersA

List all solution folders in a category.

Args: category_id: The parent category ID. page: Page number. per_page: Results per page.

Inputs: category_id (required), page, per_page. Returns: result

get_ticket_by_idA

Retrieve a single ticket by ID, including conversations and stats.

Args: ticket_id: The numeric ticket ID.

Inputs: ticket_id (required). Returns: result

get_ticket_fieldsA

Retrieve all ticket field definitions (built-in and custom).

Inputs: no named parameters. Returns: result

get_ticketsA

List all tickets with pagination.

Args: page: Page number (starts at 1). per_page: Results per page (max 100).

Inputs: page, per_page. Returns: result

get_vendorsA

List all vendors.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

get_workspacesB

List all workspaces.

Args: page: Page number. per_page: Results per page.

Inputs: page, per_page. Returns: result

list_ticket_conversationsA

List all conversations (replies and notes) for a ticket.

Args: ticket_id: The ticket ID. page: Page number. per_page: Results per page (max 100).

Inputs: ticket_id (required), page, per_page. Returns: result

reply_to_ticketA

Reply to a ticket (visible to requester).

Args: ticket_id: The ticket ID. body: Reply body (HTML supported). cc_emails: Optional CC email addresses.

Inputs: body (required), ticket_id (required), cc_emails. Returns: result

search_assetsA

Search assets by various fields.

Query examples: "name:'Dell Laptop'" "asset_tag:'ASSET-001'" "serial_number:'SN123'" "asset_state:'In Use'"

Args: query: Asset search query. page: Page number. per_page: Results per page.

Inputs: query (required), page, per_page. Returns: result

search_solution_articlesB

Search solution articles by keyword.

Args: search_term: Text to search for in articles.

Inputs: search_term (required). Returns: result

update_assetB

Update an existing asset.

Args: display_id: The asset display ID. updates: Dictionary of fields to update.

Inputs: updates (required), display_id (required). Returns: result

update_changeA

Update an existing change request.

Args: change_id: The change ID. change_fields: Dictionary of fields to update. Common fields: status, priority, impact, risk, subject, description, group_id, etc.

Inputs: change_id (required), change_fields (required). Returns: result

update_ticketA

Update an existing ticket.

Args: ticket_id: The ticket ID to update. updates: Dictionary of fields to update. Common fields: status, priority, group_id, responder_id, subject, description, category, tags, custom_fields, etc.

Inputs: updates (required), ticket_id (required). Returns: result

Tool Schema Changelog

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

Loading tool schema changelog…

    TDQS

    C2.9/5.0

    Scored across 70 tools

    Disambiguation3/5

    Most tools are clearly separated by entity type, but there is notable overlap between list, filter, and search tools for the same entities (e.g., get_assets vs search_assets vs filter_assets, get_changes vs filter_changes). An agent may struggle to choose the right variant.

    Naming Consistency5/5

    All tools use a consistent snake_case verb_noun pattern (get_, create_, update_, delete_, filter_, search_). Even variations like get_ticket_by_id or get_agent_group_by_id are predictable and follow the same convention.

    Tool Count1/5

    With 70 tools, this is far beyond the 50+ threshold for 'extreme mismatch'. The server tries to cover every Freshservice module, resulting in an overwhelming surface that is unlikely to be navigated efficiently by agents.

    Completeness2/5

    Tickets, changes, and assets have solid CRUD coverage, but many other entities (agents, groups, products, contracts, etc.) are read-only with no create/update/delete operations. This creates significant gaps where agents can retrieve data but cannot complete lifecycle workflows.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      C
      quality
      D
      maintenance
      Enables Claude to interact with ServiceNow instances through comprehensive API integration. Supports incident management, service catalog operations, change requests, knowledge base management, user administration, and agile project management with multiple authentication methods.
      82
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI models to interact with Freshservice IT service management platform, allowing automated ticket management, change requests, asset tracking, and solution article operations through natural language commands.
      MIT
    • A
      license
      B
      quality
      D
      maintenance
      Enables AI assistants to connect with Freshservice ITSM for managing tickets, assets, agents, and organizational data through natural language. It provides a comprehensive set of tools for performing CRUD operations on service desk records and searching across the Freshservice platform.
      53
      293 npm
      1
      MIT