Rollbar MCP Server
Rollbar MCP 服务器
用于 Rollbar API 集成的动态 MCP 服务器实现,使 LLM 能够与 Rollbar 错误跟踪数据进行交互。
特征
列出并过滤错误项
获取详细错误信息
查看错误发生
访问项目和环境详细信息
跟踪部署
列出用户和团队
Related MCP server: agentvet-mcp
配置
环境变量
ROLLBAR_PROJECT_TOKEN:Rollbar 项目访问令牌 - API 需要此令牌来检索项目错误和部署信息ROLLBAR_ACCOUNT_TOKEN:Rollbar 帐户访问令牌 - API 需要此令牌才能访问帐户范围内的项目和用户信息ROLLBAR_PROJECT_ID:默认项目 ID(请求中未指定时使用)- 可选ROLLBAR_PROJECT_NAME:用于参考的默认项目名称 - 可选
注意:根据您使用的功能,您可能需要
ROLLBAR_PROJECT_TOKEN、ROLLBAR_ACCOUNT_TOKEN或两者兼而有之。为了获得完整功能,建议同时配置这两个令牌,但服务只会使用特定 API 对应的令牌。
所需Token及API对应表
API | 必需令牌 |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_PROJECT_TOKEN |
| ROLLBAR_ACCOUNT_TOKEN |
| ROLLBAR_ACCOUNT_TOKEN |
| ROLLBAR_ACCOUNT_TOKEN |
| ROLLBAR_ACCOUNT_TOKEN |
您可以通过以下方式获取 Rollbar 访问令牌:
登录您的 Rollbar 帐户( https://rollbar.com/ )
对于项目令牌:设置 -> 项目访问令牌(用于项目级访问)
对于帐户令牌:设置 -> 帐户访问令牌(用于帐户级别访问)
创建具有“读取”范围的新令牌
如何使用
光标集成
添加到您的~/.cursor/mcp.json :
{
"mcpServers": {
"rollbar-mcp": {
"command": "npx",
"args": ["-y", "@hiyorineko/mcp-rollbar-server"],
"env": {
"ROLLBAR_PROJECT_TOKEN": "YOUR_PROJECT_ACCESS_TOKEN",
"ROLLBAR_ACCOUNT_TOKEN": "YOUR_ACCOUNT_ACCESS_TOKEN",
"ROLLBAR_PROJECT_ID": "YOUR_PROJECT_ID",
"ROLLBAR_PROJECT_NAME": "YOUR_PROJECT_NAME"
}
}
}
}本地
克隆此存储库后,请按照以下步骤设置 MCP 客户端:
$ cd mcp-rollbar-server
$ npm install
$ npm run build添加到您的~/.cursor/mcp.json :
{
"mcpServers": {
"rollbar-mcp": {
"command": "YOUR_NODE_PATH",
"args": ["YOUR_PROJECT_PATH/mcp-rollbar-server/dist/src/index.js"],
"env": {
"ROLLBAR_PROJECT_TOKEN": "YOUR_PROJECT_ACCESS_TOKEN",
"ROLLBAR_ACCOUNT_TOKEN": "YOUR_ACCOUNT_ACCESS_TOKEN",
"ROLLBAR_PROJECT_ID": "YOUR_PROJECT_ID",
"ROLLBAR_PROJECT_NAME": "YOUR_PROJECT_NAME"
}
}
}
}要找到“YOUR_NODE_PATH”的值,请在终端中运行哪个节点。
使用示例
List the most recent errors in my production environment.查看错误详细信息
Get detailed information for error item with ID 12345, including stack trace and recent occurrences.跟踪部署
Show me the recent deployments for project 67890.按级别过滤错误
List all critical errors that occurred in the last week.工具
rollbar_list_items
列出 Rollbar 中的项目(错误)
输入:
status(字符串,可选):按状态过滤(活动、已解决、静音等)level(字符串,可选):按级别过滤(严重、错误、警告、信息、调试)environment(字符串,可选):按环境过滤(生产、暂存等)limit(数字,可选):返回的最大项目数(默认值:20)page(数字,可选):分页的页码(默认值:1)
返回:错误项列表,其中包含计数器、级别、总发生次数等详细信息。
rollbar_get_item
使用 Rollbar 系统维护的内部项目 ID 从 Rollbar 获取特定项目(错误)。
输入:
id(数字):商品 ID
返回:有关特定错误项的详细信息
rollbar_get_item_by_counter
从 Rollbar 获取按项目计数器指定的项目。计数器是 Rollbar 界面中显示的可见 ID。
输入:
counter(数字):该项目的项目计数器
返回:有关由项目计数器标识的特定错误项的详细信息
rollbar_list_occurrences
列出 Rollbar 中出现的错误
输入:
itemId(数字,可选):用于过滤出现的项目 IDlimit(数字,可选):返回的最大出现次数(默认值:20)page(数字,可选):分页的页码(默认值:1)
返回:错误发生列表及详细信息
rollbar_get_occurrence
从 Rollbar 获取错误的具体发生情况
输入:
id(字符串):发生 ID
返回:有关特定错误发生的详细信息
rollbar_list_projects
列出 Rollbar 中的项目
输入:无
返回:项目列表及其 ID、名称和状态
rollbar_get_project
从 Rollbar 获取特定项目
输入:
id(数字):项目ID
返回:有关特定项目的详细信息
rollbar_list_environments
列出 Rollbar 中的环境
输入:
projectId(数字):项目 ID
返回:指定项目的环境列表
rollbar_list_users
列出 Rollbar 中的用户
输入:无
返回:用户列表及其 ID、用户名、电子邮件和访问级别
rollbar_get_user
从 Rollbar 获取特定用户
输入:
id(数字):用户ID
返回:有关特定用户的详细信息
rollbar_list_deploys
列出 Rollbar 的部署
输入:
projectId(数字):项目 IDenvironment(字符串,可选):环境名称limit(数字,可选):返回的最大部署数量(默认值:20)page(数字,可选):分页的页码(默认值:1)
返回:指定项目和环境的部署列表
rollbar_get_deploy
从 Rollbar 获取特定部署
输入:
deployId(数字):部署 ID
返回:有关特定部署的详细信息
Available Tools
13 toolsrollbar_get_deployC
Get a specific deploy from Rollbar
Inputs: deployId (required).
rollbar_get_itemB
Get a specific item (error) from Rollbar using the internal item ID maintained by Rollbar's system.
Inputs: id (required).
rollbar_get_item_by_counterA
Get a specific item by project counter from Rollbar. The counter is the visible ID that appears in the Rollbar UI.
Inputs: counter (required).
rollbar_get_item_by_occurrence_uuidA
Get a specific item (error) from Rollbar using an occurrence UUID. The UUID must be from an occurrence that belongs to the item.
Inputs: uuid (required).
rollbar_get_occurrenceC
Get a specific occurrence of an error from Rollbar
Inputs: id (required).
rollbar_get_projectC
Get a specific project from Rollbar
Inputs: id (required).
rollbar_get_userC
Get a specific user from Rollbar
Inputs: id (required).
rollbar_list_deploysC
List deploys from Rollbar
Inputs: projectId (required), environment, limit, page.
rollbar_list_environmentsC
List environments from Rollbar
Inputs: projectId (required).
rollbar_list_itemsC
List items (errors) from Rollbar
Inputs: status, level, environment, limit, +1 more.
rollbar_list_occurrencesC
List occurrences of errors from Rollbar
Inputs: itemId, limit, page.
rollbar_list_projectsB
List projects from Rollbar
Inputs: no named parameters.
rollbar_list_usersB
List users from Rollbar
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 13 tools
Every tool has a clearly distinct purpose with no ambiguity. The tools are organized around specific Rollbar resources (deploys, items, occurrences, projects, users, environments) with clear get vs list operations. Even the three item retrieval methods (by internal ID, counter, or occurrence UUID) are explicitly differentiated in their descriptions.
Perfectly consistent naming pattern throughout all 13 tools. Every tool follows the exact same 'rollbar_verb_noun' structure with consistent snake_case. The verbs (get/list) are used predictably across all resource types.
13 tools is well-scoped for a Rollbar integration server. Each tool earns its place by covering distinct operations across multiple resource types. The count provides comprehensive coverage without being overwhelming for the error monitoring domain.
Excellent coverage of read operations for all core Rollbar resources with consistent get/list pairs. Minor gaps exist in write operations (no create/update/delete tools), but this appears intentional for a monitoring-focused server. Agents can perform most common error investigation workflows.
Maintenance
Related MCP Connectors
MCP server for Sentry - error monitoring, issue tracking, and debugging for AI assistants
An MCP server that provides an API to LLMs to manage their JumpCloud resources.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server for AI access to SmartBear tools, including BugSnag, Reflect, Swagger, PactFlow, QTM4J.
Related MCP Servers
- FlicenseNot gradedqualityCmaintenanceAn MCP server that enables AI agents to retrieve and analyze error reports from a self-hosted Telebugs instance. It provides tools for listing projects, searching error groups, and fetching detailed reports directly from the Telebugs SQLite database.4-
- AlicenseAqualityDmaintenanceMCP server that validates LLM-generated tool-call arguments, lints tool definitions, and produces retry messages for AI assistants.346 npm1MIT
- AlicenseNot gradedqualityDmaintenanceMCP server that provides tools to interact with the LiteLLM proxy API, enabling LLM completions, embeddings, image generation, and admin operations.MIT
- AlicenseAqualityDmaintenanceMCP server for integrating self-hosted Sentry with AI assistants, enabling project and issue listing, issue details with stack traces, and event retrieval.710 npm1MIT