AI 도우미가 Intercom의 고객 지원 데이터에 액세스하여 분석할 수 있도록 하는 MCP 호환 서버입니다.
고급 필터링으로 대화 및 티켓 검색
고객, 상태, 날짜 범위 및 키워드로 필터링
연락처가 없어도 이메일 내용으로 검색
Intercom의 검색 API를 통한 효율적인 서버 측 필터링
MCP 호환 AI 어시스턴트와의 원활한 통합
Related MCP server: jitbit-helpdesk-mcp
Node.js 18.0.0 이상
API 액세스가 가능한 Intercom 계정
Intercom API 토큰(Intercom 계정 설정에서 사용 가능)
지엑스피1
기본 Docker 구성은 Glama 호환성을 위해 최적화되었습니다.
# Start Docker (if not already running)
# On Windows: Start Docker Desktop application
# On Linux: sudo systemctl start docker
# Build the image
docker build -t mcp-intercom .
# Run the container with your API token and port mappings
docker run --rm -it -p 3000:3000 -p 8080:8080 -e INTERCOM_ACCESS_TOKEN="your_token_here" mcp-intercom:latest검증 단계:
# Test the server status
curl -v http://localhost:8080/.well-known/glama.json
# Test the MCP endpoint
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1,"method":"mcp.capabilities"}' http://localhost:3000Glama 특정 종속성이 없는 더 가벼운 버전을 선호하는 경우:
# Build the standard image
docker build -t mcp-intercom-standard -f Dockerfile.standard .
# Run the standard container
docker run --rm -it -p 3000:3000 -p 8080:8080 -e INTERCOM_ACCESS_TOKEN="your_token_here" mcp-intercom-standard:latest기본 버전에는 Glama 플랫폼과 통합하는 데 필요한 특정 종속성과 구성이 포함되어 있는 반면, 표준 버전은 더 가볍습니다.
list_conversations콘텐츠 필터링을 통해 특정 날짜 범위 내의 모든 대화를 검색합니다.
매개변수:
startDate (DD/MM/YYYY) – 시작일(필수)
endDate (DD/MM/YYYY) – 종료 날짜(필수)
keyword (문자열) – 이 텍스트가 포함된 대화를 포함하도록 필터링합니다.
exclude (문자열) – 이 텍스트가 포함된 대화를 제외하기 위한 필터
참고사항:
날짜 범위는 7일을 초과할 수 없습니다.
Intercom의 검색 API를 통해 효율적인 서버 측 필터링을 사용합니다.
예:
{
"startDate": "15/01/2025",
"endDate": "21/01/2025",
"keyword": "billing"
}search_conversations_by_customer특정 고객에 대한 대화를 찾습니다.
매개변수:
customerIdentifier (문자열) – 고객 이메일 또는 Intercom ID(필수)
startDate (DD/MM/YYYY) – 선택 시작일
endDate (DD/MM/YYYY) – 선택적인 종료 날짜
keywords (배열) – 콘텐츠로 필터링할 선택적 키워드
참고사항:
연락처가 없더라도 이메일 내용으로 대화를 찾을 수 있습니다.
효율적인 검색을 위해 이메일을 연락처 ID로 변환합니다.
예:
{
"customerIdentifier": "customer@example.com",
"startDate": "15/01/2025",
"endDate": "21/01/2025",
"keywords": ["billing", "refund"]
}search_tickets_by_status티켓을 상태별로 검색합니다.
매개변수:
status (문자열) – "열림", "보류 중" 또는 "해결됨"(필수)
startDate (DD/MM/YYYY) – 선택 시작일
endDate (DD/MM/YYYY) – 선택적인 종료 날짜
예:
{
"status": "open",
"startDate": "15/01/2025",
"endDate": "21/01/2025"
}search_tickets_by_customer특정 고객과 관련된 티켓을 찾습니다.
매개변수:
customerIdentifier (문자열) – 고객 이메일 또는 Intercom ID(필수)
startDate (DD/MM/YYYY) – 선택 시작일
endDate (DD/MM/YYYY) – 선택적인 종료 날짜
예:
{
"customerIdentifier": "customer@example.com",
"startDate": "15/01/2025",
"endDate": "21/01/2025"
}claude_desktop_config.json 에 다음을 추가하세요:
{
"mcpServers": {
"intercom-mcp": {
"command": "intercom-mcp",
"args": [],
"env": {
"INTERCOM_ACCESS_TOKEN": "your_intercom_api_token"
}
}
}
}이 서버가 Intercom API와 통합되는 방식에 대한 자세한 기술 정보는 src/services/INTERCOM_API_NOTES.md 참조하세요. 이 문서에서는 매개변수 매핑, Intercom 엔드포인트 사용법, 그리고 개발자를 위한 구현 세부 정보를 설명합니다.
# Clone and install dependencies
git clone https://github.com/raoulbia-ai/mcp-server-for-intercom.git
cd mcp-server-for-intercom
npm install
# Build and run for development
npm run build
npm run dev
# Run tests
npm test이 프로젝트는 독립적인 통합 프로젝트이며 Intercom Inc.와 제휴 관계가 없고, 공식적으로 연결되거나 보증을 받지 않았습니다. "Intercom"은 Intercom Inc.의 등록 상표입니다.
이 프로젝트는 Apache License 2.0에 따라 라이선스가 부여되었습니다. 자세한 내용은 LICENSE 파일을 참조하세요.
Retrieves Intercom conversations within a specific date range.
Required: startDate, endDate (DD/MM/YYYY format, max 7-day range) Optional: keyword, exclude (for content filtering)
Always ask for specific dates when user makes vague time references.
| Name | Required | Description | Default |
|---|---|---|---|
| endDate | Yes | End date in DD/MM/YYYY format (e.g., '21/01/2025'). Required. | |
| exclude | No | Optional exclusion filter for conversation content. | |
| keyword | No | Optional keyword to filter conversations by content. | |
| startDate | Yes | Start date in DD/MM/YYYY format (e.g., '15/01/2025'). Required. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It adds some context beyond basic functionality: it specifies the date format (DD/MM/YYYY), max range (7 days), and that parameters are required or optional. However, it doesn't cover important behavioral aspects like rate limits, authentication needs, pagination, or what the return format looks like (especially since there's no output schema). For a tool with no annotations, this leaves gaps in transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded: the first sentence states the core purpose, followed by details on parameters and a usage directive. Each sentence adds value, with no wasted words. It could be slightly more structured (e.g., bullet points for parameters), but it's efficient and clear.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (a retrieval tool with 4 parameters), no annotations, and no output schema, the description is somewhat complete but has gaps. It covers the basic purpose, parameter requirements, and a usage tip, but lacks details on behavioral traits (e.g., rate limits), output format, and how it differs from siblings. For a tool without structured support, it should do more to compensate.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, meaning the input schema already documents all parameters thoroughly. The description adds minimal value beyond the schema: it reiterates that startDate and endDate are required and in DD/MM/YYYY format, and mentions the max 7-day range (which isn't in the schema). However, it doesn't provide additional semantic context for the optional parameters (keyword, exclude) or explain their interactions. Baseline 3 is appropriate when the schema does most of the work.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Retrieves Intercom conversations within a specific date range.' This specifies the verb ('retrieves'), resource ('Intercom conversations'), and scope ('within a specific date range'). However, it doesn't explicitly differentiate from sibling tools like 'search_conversations_by_customer,' which appears to be a more targeted search tool, so it doesn't reach the highest score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context for usage: it specifies that the tool is for retrieving conversations within a date range and includes a directive to 'Always ask for specific dates when user makes vague time references.' This offers practical guidance on when to use it (for date-based retrieval) and how to handle ambiguous inputs. However, it doesn't explicitly state when not to use it or mention alternatives like the sibling tools, so it's not a perfect 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Searches for conversations by customer email or ID with optional date filtering.
Required: customerIdentifier (email/ID) Optional: startDate, endDate (DD/MM/YYYY format) Optional: keywords (array of terms to filter by)
Use when looking for conversation history with a specific customer.
| Name | Required | Description | Default |
|---|---|---|---|
| customerIdentifier | Yes | Customer email or ID to search for | |
| endDate | No | Optional end date in DD/MM/YYYY format (e.g., '21/01/2025') | |
| keywords | No | Optional keywords to filter conversations by content | |
| startDate | No | Optional start date in DD/MM/YYYY format (e.g., '15/01/2025') |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It mentions optional date filtering and keywords, but lacks critical behavioral details like whether this is a read-only operation, what permissions are needed, how results are returned (e.g., pagination), or error handling. For a search tool with no annotation coverage, this is a significant gap.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Perfectly concise and well-structured: first sentence states purpose, bullet points clarify parameters, and final sentence provides usage guidance. Every sentence earns its place with zero waste, and information is front-loaded appropriately.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations and no output schema, the description is incomplete for a search tool. It covers purpose and parameters well, but misses behavioral aspects like result format, pagination, or error cases. However, it's adequate for basic usage, so it meets the minimum viable threshold.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters thoroughly. The description adds minimal value by restating that customerIdentifier is required and dates/keywords are optional, but doesn't provide additional context beyond what's in the schema. Baseline 3 is appropriate when schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with a specific verb ('searches') and resource ('conversations'), and distinguishes it from siblings by specifying it searches by customer email/ID rather than listing all conversations or searching tickets. The title being null doesn't affect this clarity.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly states when to use this tool ('Use when looking for conversation history with a specific customer'), which differentiates it from sibling tools like list_conversations (general listing) and search_tickets_by_customer (ticket-focused). No when-not guidance, but the context is clear enough for full credit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Searches for tickets by customer email or ID with optional date filtering.
Required: customerIdentifier (email/ID) Optional: startDate, endDate (DD/MM/YYYY format)
Use when analyzing a customer's support history.
| Name | Required | Description | Default |
|---|---|---|---|
| customerIdentifier | Yes | Customer email or ID to search for | |
| endDate | No | Optional end date in DD/MM/YYYY format (e.g., '21/01/2025') | |
| startDate | No | Optional start date in DD/MM/YYYY format (e.g., '15/01/2025') |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. While it mentions the tool is for searching (implying read-only), it doesn't disclose important behavioral traits like whether this is a safe read operation, if it requires specific permissions, rate limits, pagination behavior, or what the response format looks like. The description adds minimal behavioral context beyond the basic function.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized with three sentences that are front-loaded with the core purpose. Each sentence earns its place by providing purpose, parameter guidance, and usage context. Minor improvements could include combining the parameter details into a single sentence for better flow.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity (3 parameters, no output schema, no annotations), the description is somewhat complete but has gaps. It covers the purpose and basic usage but lacks details on behavioral traits, response format, and explicit differentiation from sibling tools. The absence of annotations and output schema increases the need for more comprehensive description, which isn't fully met.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already fully documents all parameters. The description adds some value by explicitly labeling parameters as 'Required' and 'Optional' and specifying the date format, but this information is largely redundant with the schema. The baseline score of 3 reflects adequate but not exceptional added semantic value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool searches for tickets by customer email or ID with optional date filtering, providing a specific verb ('searches') and resource ('tickets'). It distinguishes from sibling tools like 'search_tickets_by_status' by specifying customer-based search, though it doesn't explicitly mention how it differs from 'search_conversations_by_customer' beyond the resource type.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context for when to use the tool ('when analyzing a customer's support history'), which helps guide usage. However, it doesn't explicitly state when NOT to use it or mention specific alternatives among the sibling tools, such as when to choose 'search_conversations_by_customer' instead.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Searches for tickets by status with optional date filtering.
Required: status (one of: open, pending, resolved) Optional: startDate, endDate (DD/MM/YYYY format)
Use when analyzing support workload or tracking issue resolution.
| Name | Required | Description | Default |
|---|---|---|---|
| endDate | No | Optional end date in DD/MM/YYYY format (e.g., '21/01/2025') | |
| startDate | No | Optional start date in DD/MM/YYYY format (e.g., '15/01/2025') | |
| status | Yes | Ticket status to search for (open, pending, or resolved) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. While it mentions the search functionality and date format, it lacks critical behavioral details: it doesn't specify whether this is a read-only operation, what permissions might be required, how results are returned (e.g., pagination, format), or any rate limits. For a search tool with no annotation coverage, this leaves significant gaps in understanding its behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is efficiently structured with three sentences: purpose statement, parameter requirements, and usage guidelines. Each sentence adds distinct value without redundancy. It's appropriately sized and front-loaded with the core functionality.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations and no output schema, the description is moderately complete for a search tool. It covers the basic purpose, parameters, and usage context, but lacks details about behavioral aspects (e.g., read-only nature, result format, error handling) and doesn't fully address sibling tool differentiation. For a tool with 3 parameters and no structured safety hints, more behavioral context would be beneficial.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema description coverage is 100%, with all parameters well-documented in the schema (status with enum values, startDate/endDate with format). The description adds minimal value beyond the schema: it repeats the status options and date format, but doesn't provide additional context like how date filtering interacts with status or example use cases for the parameters. This meets the baseline for high schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Searches for tickets by status with optional date filtering.' This specifies the verb ('searches'), resource ('tickets'), and scope ('by status with optional date filtering'). However, it doesn't explicitly differentiate from sibling tools like 'search_tickets_by_customer' or 'search_conversations_by_customer', which would require mentioning customer vs. status filtering.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear usage context: 'Use when analyzing support workload or tracking issue resolution.' This gives practical scenarios for when to use the tool. However, it doesn't explicitly state when NOT to use it or mention alternatives like the sibling tools, which would be needed for a perfect score.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0list_conversationssearch_conversations_by_customersearch_tickets_by_customersearch_tickets_by_statusScored across 4 tools
The tools are mostly distinct with clear purposes: list_conversations retrieves all conversations in a date range, while the other three are specific searches (by customer for conversations/tickets, by status for tickets). However, search_conversations_by_customer and search_tickets_by_customer could be slightly confused since both target customers, but their descriptions clarify the resource difference (conversations vs. tickets).
All tool names follow a consistent verb_noun pattern with snake_case: list_conversations, search_conversations_by_customer, search_tickets_by_customer, search_tickets_by_status. The naming is predictable and readable throughout the set.
With only 4 tools, the set feels thin for an Intercom server, which typically handles a broader range of operations like creating/updating conversations, managing contacts, or sending messages. While the tools cover basic retrieval and search, the count is borderline low for the domain's potential scope.
The tool surface is significantly incomplete for an Intercom integration. It only provides search and list operations, missing essential CRUD actions like create_conversation, update_ticket, or delete operations. There are also gaps in managing other Intercom resources such as contacts, companies, or messages, which will limit agent capabilities.
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
Build and manage AI-native customer support agents from Claude or any MCP client.
MCP server that lets AI assistants use all OneSchema features exposed via the public API.
Cloud-hosted MCP server for secure AI access to enterprise data sources via CData Connect AI.
