本文へ移動
cccskills
無料GitHub で公開

freshdesk-automation

Automate Freshdesk helpdesk operations including tickets, contacts, companies, notes, and replies via Rube MCP (Composio). Always search tools first for current schemas.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md11.3 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Freshdesk Automation via Rube MCP

Automate Freshdesk customer support workflows including ticket management, contact and company operations, notes, replies, and ticket search through Composio's Freshdesk toolkit.

Toolkit docs: composio.dev/toolkits/freshdesk

Prerequisites

  • Rube MCP must be connected (RUBE_SEARCH_TOOLS available)
  • Active Freshdesk connection via RUBE_MANAGE_CONNECTIONS with toolkit freshdesk
  • Always call RUBE_SEARCH_TOOLS first to get current tool schemas

Setup

Get Rube MCP: Add https://rube.app/mcp as an MCP server in your client configuration. No API keys needed — just add the endpoint and it works.

  1. Verify Rube MCP is available by confirming RUBE_SEARCH_TOOLS responds
  2. Call RUBE_MANAGE_CONNECTIONS with toolkit freshdesk
  3. If connection is not ACTIVE, follow the returned auth link to complete Freshdesk authentication
  4. Confirm connection status shows ACTIVE before running any workflows

Core Workflows

1. Create and Manage Tickets

When to use: User wants to create a new support ticket, update an existing ticket, or view ticket details.

Tool sequence:

  1. FRESHDESK_SEARCH_CONTACTS - Find requester by email to get requester_id [Optional]
  2. FRESHDESK_LIST_TICKET_FIELDS - Check available custom fields and statuses [Optional]
  3. FRESHDESK_CREATE_TICKET - Create a new ticket with subject, description, requester info [Required]
  4. FRESHDESK_UPDATE_TICKET - Modify ticket status, priority, assignee, or other fields [Optional]
  5. FRESHDESK_VIEW_TICKET - Retrieve full ticket details by ID [Optional]

Key parameters for FRESHDESK_CREATE_TICKET:

  • subject: Ticket subject (required)
  • description: HTML content of the ticket (required)
  • email: Requester email (at least one requester identifier required)
  • requester_id: User ID of requester (alternative to email)
  • status: 2=Open, 3=Pending, 4=Resolved, 5=Closed (default 2)
  • priority: 1=Low, 2=Medium, 3=High, 4=Urgent (default 1)
  • source: 1=Email, 2=Portal, 3=Phone, 7=Chat (default 2)
  • responder_id: Agent ID to assign the ticket to
  • group_id: Group to assign the ticket to
  • tags: Array of tag strings
  • custom_fields: Object with cf_<field_name> keys

Pitfalls:

  • At least one requester identifier is required: requester_id, email, phone, facebook_id, twitter_id, or unique_external_id
  • If phone is provided without email, then name becomes mandatory
  • description supports HTML formatting
  • attachments field expects multipart/form-data format, not file paths or URLs
  • Custom field keys must be prefixed with cf_ (e.g., cf_reference_number)
  • Status and priority are integers, not strings

2. Search and Filter Tickets

When to use: User wants to find tickets by status, priority, date range, agent, or custom fields.

Tool sequence:

  1. FRESHDESK_GET_TICKETS - List tickets with simple filters (status, priority, agent) [Required]
  2. FRESHDESK_GET_SEARCH - Advanced ticket search with query syntax [Required]
  3. FRESHDESK_VIEW_TICKET - Get full details for specific tickets from results [Optional]
  4. FRESHDESK_LIST_TICKET_FIELDS - Check available fields for search queries [Optional]

Key parameters for FRESHDESK_GET_TICKETS:

  • status: Filter by status integer (2=Open, 3=Pending, 4=Resolved, 5=Closed)
  • priority: Filter by priority integer (1-4)
  • agent_id: Filter by assigned agent
  • requester_id: Filter by requester
  • email: Filter by requester email
  • created_since: ISO 8601 timestamp
  • page / per_page: Pagination (default 30 per page)
  • sort_by / sort_order: Sort field and direction

Key parameters for FRESHDESK_GET_SEARCH:

  • query: Query string like "status:2 AND priority:3" or "(created_at:>'2024-01-01' AND tag:'urgent')"
  • page: Page number (1-10, max 300 total results)

Pitfalls:

  • FRESHDESK_GET_SEARCH query must be enclosed in double quotes
  • Query string limited to 512 characters
  • Maximum 10 pages (300 results) from search endpoints
  • Date fields in queries use UTC format YYYY-MM-DD
  • Use null keyword to find tickets with empty fields (e.g., "agent_id:null")
  • FRESHDESK_LIST_ALL_TICKETS takes no parameters and returns all tickets (use GET_TICKETS for filtering)

3. Reply to and Add Notes on Tickets

When to use: User wants to send a reply to a customer, add internal notes, or view conversation history.

Tool sequence:

  1. FRESHDESK_VIEW_TICKET - Verify ticket exists and check current state [Prerequisite]
  2. FRESHDESK_REPLY_TO_TICKET - Send a public reply to the requester [Required]
  3. FRESHDESK_ADD_NOTE_TO_TICKET - Add a private or public note [Required]
  4. FRESHDESK_LIST_ALL_TICKET_CONVERSATIONS - View all messages and notes on a ticket [Optional]
  5. FRESHDESK_UPDATE_CONVERSATIONS - Edit an existing note [Optional]

Key parameters for FRESHDESK_REPLY_TO_TICKET:

  • ticket_id: Ticket ID (integer, required)
  • body: Reply content, supports HTML (required)
  • cc_emails / bcc_emails: Additional recipients (max 50 total across to/cc/bcc)
  • from_email: Override sender email if multiple support emails configured
  • user_id: Agent ID to reply on behalf of

Key parameters for FRESHDESK_ADD_NOTE_TO_TICKET:

  • ticket_id: Ticket ID (integer, required)
  • body: Note content, supports HTML (required)
  • private: true for agent-only visibility, false for public (default true)
  • notify_emails: Only accepts agent email addresses, not external contacts

Pitfalls:

  • There are two reply tools: FRESHDESK_REPLY_TO_TICKET (more features) and FRESHDESK_REPLY_TICKET (simpler); both work
  • FRESHDESK_ADD_NOTE_TO_TICKET defaults to private (agent-only); set private: false for public notes
  • notify_emails in notes only accepts agent emails, not customer emails
  • Only notes can be edited via FRESHDESK_UPDATE_CONVERSATIONS; incoming replies cannot be edited

4. Manage Contacts and Companies

When to use: User wants to create, search, or manage customer contacts and company records.

Tool sequence:

  1. FRESHDESK_SEARCH_CONTACTS - Search contacts by email, phone, or company [Required]
  2. FRESHDESK_GET_CONTACTS - List contacts with filters [Optional]
  3. FRESHDESK_IMPORT_CONTACT - Bulk import contacts from CSV [Optional]
  4. FRESHDESK_SEARCH_COMPANIES - Search companies by custom fields [Required]
  5. FRESHDESK_GET_COMPANIES - List all companies [Optional]
  6. FRESHDESK_CREATE_COMPANIES - Create a new company [Optional]
  7. FRESHDESK_UPDATE_COMPANIES - Update company details [Optional]
  8. FRESHDESK_LIST_COMPANY_FIELDS - Check available company fields [Optional]

Key parameters for FRESHDESK_SEARCH_CONTACTS:

  • query: Search string like "email:'user@example.com'" (required)
  • page: Pagination (1-10, max 30 per page)

Key parameters for FRESHDESK_CREATE_COMPANIES:

  • name: Company name (required)
  • domains: Array of domain strings for auto-association with contacts
  • health_score: "Happy", "Doing okay", or "At risk"
  • account_tier: "Basic", "Premium", or "Enterprise"
  • industry: Standard industry classification

Pitfalls:

  • FRESHDESK_SEARCH_CONTACTS requires exact matches; partial/regex searches are not supported
  • FRESHDESK_SEARCH_COMPANIES cannot search by standard name field; use custom fields or created_at
  • Company custom fields do NOT use the cf_ prefix (unlike ticket custom fields)
  • domains on companies enables automatic contact-to-company association by email domain
  • Contact search queries require string values in single quotes inside double-quoted query

Common Patterns

ID Resolution

Always resolve display values to IDs before operations:

  • Requester email -> requester_id: FRESHDESK_SEARCH_CONTACTS with "email:'user@example.com'"
  • Company name -> company_id: FRESHDESK_GET_COMPANIES and match by name (search by name not supported)
  • Agent name -> agent_id: Not directly available; use agent_id from ticket responses or admin configuration

Pagination

Freshdesk uses page-based pagination:

  • FRESHDESK_GET_TICKETS: page (starting at 1) and per_page (max 100)
  • FRESHDESK_GET_SEARCH: page (1-10, 30 results per page, max 300 total)
  • FRESHDESK_SEARCH_CONTACTS: page (1-10, 30 per page)
  • FRESHDESK_LIST_ALL_TICKET_CONVERSATIONS: page and per_page (max 100)

Known Pitfalls

ID Formats

  • Ticket IDs, contact IDs, company IDs, agent IDs, and group IDs are all integers
  • There are no string-based IDs in Freshdesk

Rate Limits

  • Freshdesk enforces per-account API rate limits based on plan tier
  • Bulk operations should be paced to avoid 429 responses
  • Search endpoints are limited to 300 total results (10 pages of 30)

Parameter Quirks

  • Status values: 2=Open, 3=Pending, 4=Resolved, 5=Closed (integers, not strings)
  • Priority values: 1=Low, 2=Medium, 3=High, 4=Urgent (integers, not strings)
  • Source values: 1=Email, 2=Portal, 3=Phone, 7=Chat, 9=Feedback Widget, 10=Outbound Email
  • Ticket custom fields use cf_ prefix; company custom fields do NOT
  • description in tickets supports HTML formatting
  • Search query strings must be in double quotes with string values in single quotes
  • FRESHDESK_LIST_ALL_TICKETS returns all tickets with no filter parameters

Response Structure

  • Ticket details include nested objects for requester, assignee, and conversation data
  • Search results are paginated with a maximum of 300 results across 10 pages
  • Conversation lists include both replies and notes in chronological order

Quick Reference

TaskTool SlugKey Params
Create ticketFRESHDESK_CREATE_TICKETsubject, description, email, priority
Update ticketFRESHDESK_UPDATE_TICKETticket_id, status, priority
View ticketFRESHDESK_VIEW_TICKETticket_id
List ticketsFRESHDESK_GET_TICKETSstatus, priority, page, per_page
List all ticketsFRESHDESK_LIST_ALL_TICKETS(none)
Search ticketsFRESHDESK_GET_SEARCHquery, page
Reply to ticketFRESHDESK_REPLY_TO_TICKETticket_id, body, cc_emails
Reply (simple)FRESHDESK_REPLY_TICKETticket_id, body
Add noteFRESHDESK_ADD_NOTE_TO_TICKETticket_id, body, private
List conversationsFRESHDESK_LIST_ALL_TICKET_CONVERSATIONSticket_id, page
Update noteFRESHDESK_UPDATE_CONVERSATIONSconversation_id, body
Search contactsFRESHDESK_SEARCH_CONTACTSquery, page
List contactsFRESHDESK_GET_CONTACTSemail, company_id, page
Import contactsFRESHDESK_IMPORT_CONTACTfile, name_column_index, email_column_index
Create companyFRESHDESK_CREATE_COMPANIESname, domains, industry
Update companyFRESHDESK_UPDATE_COMPANIEScompany_id, name, domains
Search companiesFRESHDESK_SEARCH_COMPANIESquery, page
List companiesFRESHDESK_GET_COMPANIESpage
List ticket fieldsFRESHDESK_LIST_TICKET_FIELDS(none)
List company fieldsFRESHDESK_LIST_COMPANY_FIELDS(none)

Powered by Composio

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Automate ActiveCampaign tasks via Rube MCP (Composio): manage contacts, tags, list subscriptions, automation enrollment, and tasks. Always search tools first for current schemas.

日本語の概要は準備中です。原文の説明を表示しています。

davepoon/buildwithclaude3,6292026年10月12日 更新

Analytics your AI agent can actually use. Track, analyze, run A/B experiments, and optimize across all your projects via CLI. Includes a growth playbook so your agent knows HOW to grow, not just what to track.

日本語の概要は準備中です。原文の説明を表示しています。

davepoon/buildwithclaude3,6292026年10月12日 更新

Teaches when to recall from long-term memory before acting and when to save durable decisions, corrections and failures afterwards. Use when a memory tool or MCP memory server is connected but the agent is not using it consistently, when the user complains that the assistant forgets preferences, conventions or past decisions between sessions, or when setting up persistent memory for a project. Works with any memory backend: a folder of Markdown files, a local MCP server, or a managed service.

日本語の概要は準備中です。原文の説明を表示しています。

davepoon/buildwithclaude3,6292026年10月12日 更新

Audit local AI coding-agent sessions with agenttrace for cost, tokens, tool failures, latency, anomalies, health, diffs, and CI gates.

日本語の概要は準備中です。原文の説明を表示しています。

davepoon/buildwithclaude3,6292026年10月12日 更新

Audit whether a website can be found, crawled, and cited by AI answer engines such as ChatGPT Search, Perplexity, Google AI Overviews, and Microsoft Copilot. Use when someone asks why their brand is missing from AI answers, whether AI crawlers can read their site, how to get cited by ChatGPT or Perplexity, or asks for a GEO or AEO (generative / answer engine optimization) review. Produces a citation baseline across buyer-intent prompts, a crawler-access check, a citability review of named pages, and a ranked fix list. Not for keyword rank tracking, paid search, or pages behind a login.

日本語の概要は準備中です。原文の説明を表示しています。

davepoon/buildwithclaude3,6292026年10月12日 更新

Automate Airtable tasks via Rube MCP (Composio): records, bases, tables, fields, views. Always search tools first for current schemas.

日本語の概要は準備中です。原文の説明を表示しています。

davepoon/buildwithclaude3,6292026年10月12日 更新

davepoon のスキルをすべて見る

このスキルの問題を報告する