跳过安装。2 分钟内即可运行。
在 cloud.anythingmcp.com 开始免费试用,一键添加 Rippling,然后将你的 AI 客户端(Claude、ChatGPT、Copilot 或 Cursor)指向生成的 MCP 端点。无需 Docker,无需 git clone,无需任何工程经验。
AnythingMCP 已收录在 Claude 的连接器目录中:一键添加到 Claude (在新标签页中打开),然后登录并选择 Claude 可以使用哪些 MCP 服务器。
摘要
通过 REST API 访问 HR 平台 Rippling:员工(worker)和用户、部门、团队、工作地点、职级、职位、雇佣类型和自定义字段、薪酬、休假类型、余额和申请、超级组(supergroup)以及工时记录。它还能提交和审批休假、记录工时、维护组织架构并创建待入职草稿(draft hire)。
试试这样问
Rippling 的示例提示
点击任意提示即可复制,然后粘贴到 Claude、ChatGPT、Cursor、Gemini、Copilot 或 OpenClaw 中,在该连接器上运行。
Claude 是 AI,可能会出错。请核对回答。
💡 无需安装? 直接使用 cloud.anythingmcp.com。 登录、点击 Connectors → Rippling、粘贴凭证、生成 MCP API key,即可完成。无 Docker、无
git clone、无本地服务器。
Rippling + Claude
通过 REST API 访问 HR 平台 Rippling:员工(worker)和用户、部门、团队、工作地点、职级、职位、雇佣类型和自定义字段、薪酬、休假类型、余额和申请、超级组(supergroup)以及工时记录。它还能提交和审批休假、记录工时、维护组织架构并创建待入职草稿(draft hire)。
前提条件
完整的设置说明已内置在连接器中 (在 store 中选择连接器时可见)。所需环境变量:
RIPPLING_API_TOKEN
步骤 1 — 获取凭证
设置大约需要 5 分钟。你需要一个拥有 Create and manage API Tokens 权限、并能访问 AI 要读取的数据的 Rippling 用户。
- 在 Rippling 中打开 Developer > API credentials(
https://app.rippling.com/api-tokens/tokens),点击 New API token。 - 为令牌命名并选择它的 scopes。令牌只能访问其 scopes 和所有者权限配置同时允许的数据。按需选择:
companies.read(必选,用于安装检查),以及workers.read、users.read、departments.read、teams.read、work-locations.read、levels.read、titles.read、employment-types.read、job-codes.read、custom-fields.read、supergroups.read。- 只有当 AI 需要查看薪资时才选
compensations.read。 leave-types.read、leave-balances.read、leave-requests.read,要提交或审批休假再加leave-requests.read-write。time-entries.read,要记录工时再加time-entries.read-write(需要 Rippling Time)。- 对应的写入操作需要
departments.read-write、teams.read-write、work-locations.read-write和draft-hires.read-write。
- 令牌只显示一次,请立即复制到
RIPPLING_API_TOKEN,不要带Bearer。安装连接器:rippling_list_companies用于安装检查。
请求以令牌所有者的身份执行。Rippling 会吊销 30 天未使用的令牌,以及已离职所有者的令牌,因此请选择访问权限稳定的所有者。
个人数据:员工、用户和薪酬方案包含个人和薪资数据:姓名、私人邮箱、电话、家庭住址、出生日期、性别和薪资。只请求任务需要的字段,不要无故复制到其他系统,并且只给令牌团队实际使用的 scopes。
读取内容:公司(rippling_list_companies);员工(worker),即雇员和承包人的雇佣记录,包含状态、工作和个人邮箱、入职和离职日期、职位、部门、经理、职级、团队、地点、雇佣类型和国家(rippling_list_workers、rippling_get_worker);用户,即员工 user_id 背后的登录身份(rippling_list_users、rippling_get_user);薪酬方案,包含年薪、月薪、周薪和时薪金额、奖金目标和佣金(rippling_list_compensations,以及用员工的 compensation_id 调用 rippling_get_compensation);自定义字段定义(rippling_list_custom_fields),字段值在员工上加 expand=custom_fields 返回;由部门、团队、工作地点、职级、职位、雇佣类型和工作代码组成的组织架构;超级组及其成员;休假类型、以小时和天计的休假余额,以及休假申请;工时记录,包含工作班次、休息、时长、状态和工资周期。expand(例如 user,manager,department)可在一次调用中嵌入关联对象。
写入内容:
- 休假:
rippling_create_leave_request为员工提交休假(PENDING走正常审批流程,APPROVED记录已经商定的休假);rippling_update_leave_request通过status批准、拒绝或取消申请,也可修改日期和备注。 - 工时:
rippling_create_time_entry以带开始和结束时间的工作班次,或以某天的时长记录工作时间;它接受idempotency_key,重试时不会产生重复记录。rippling_update_time_entry修改记录;即使worker_id不变,API 也要求提供它。 - 组织架构:
rippling_create_department、rippling_update_department、rippling_create_team、rippling_update_team、rippling_create_work_location和rippling_update_work_location。更新只修改发送的字段。 - 招聘:
rippling_create_draft_hires最多创建五条待入职草稿,在正式录用前要经过 Rippling 的审核和审批。
安装时关闭:rippling_delete_team(员工会失去团队分配)和 rippling_delete_time_entry。两者都无法撤销;团队需要时由工作区管理员开启。
筛选:filter 接受针对各工具所列字段的 OData 风格表达式,例如 status eq 'ACTIVE'、updated_at ge 2026-01-01T00:00:00 或 status in ('HIRED','ACCEPTED'),可用 and 和 or 组合。字符串用单引号,一个筛选最多 64 个节点。某位员工的余额:worker_id eq '...';待审批的申请:status eq 'PENDING'。
数据格式:ID 是 24 位十六进制字符串。日期为 YYYY-MM-DD,时间戳为带偏移的 ISO 8601。金额格式为 { "currency_type": "USD", "value": 120000 },以货币的主单位表示。员工的 status 为 INIT、HIRED、ACCEPTED、ACTIVE 或 TERMINATED。令牌无权查看的字段会返回 null,并在 __meta.redacted_fields 中注明原因,所以 null 并不能说明字段为空:请先检查 scopes 和所有者的权限。
分页:列表返回 results 和 next_link,默认 50 条,用 limit 最多 100 条。只要 next_link 不为 null,AI 就把 cursor 设为 next_link 中的游标值再次调用。
限制和常见错误:每个端点和令牌每 10 秒最多 600 次请求,超出返回 429(等几秒后重试)。400 表示筛选有误、字段无法展开或 limit 超过 100;401 表示令牌错误或已吊销;403 表示缺少 scope 或所有者没有相应权限;404 表示 ID 不存在。
连接器不会编辑员工记录或薪酬,不会直接录用或解雇任何人,也不会运行薪资发放。
步骤 2 — 安装 adapter
curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d
步骤 3 — 在 Claude 中添加连接器
你的服务器 URL: 在 AnythingMCP 中打开 MCP Servers → 此连接器所在的服务器,复制它的 URL(
https://cloud.anythingmcp.com/mcp/…)。本指南中出现YOUR_SERVER_ID的地方都用它替换。
推荐路径:在 claude.ai web 上工作,无需触碰任何配置文件。
- 在 Claude 中打开 Customize → Connectors。
- 点击 +,然后点击 Add custom connector。
- 输入名称(例如
Rippling)和远程 MCP 服务器 URLhttps://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID,然后点击 Add。 - 点击 Connect,登录 AnythingMCP(OAuth)并允许访问。
下面所有工具都会出现在你的聊天中,开始输入提示。
可用工具
| Tool | What it does |
|---|---|
rippling_list_companies | List the Rippling company the token belongs to (name, legal name, primary email, address, legal entities when expanded) |
rippling_list_workers | List workers (employees and contractors): status, work and personal email, start and end date, title, department, manager, level, teams… |
rippling_get_worker | Read one worker by id with the same fields as rippling_list_workers; expand user, manager or custom_fields for more |
rippling_list_users | List users (the login identity behind a worker): name, display name, emails, phone numbers, addresses, time zone, locale and whether the… |
rippling_get_user | Read one user by id: name, emails, phone numbers, addresses, locale and time zone |
rippling_list_compensations | List compensation packages the token owner may see: annual, monthly, weekly and hourly amounts with currency, bonus targets, commission and… |
rippling_get_compensation | Read one compensation package by id (the compensation_id of a worker) |
rippling_list_custom_fields | List the company's custom fields on employees: id, name, description, data type (TEXT, DATE, NUMBER, SELECT and more), whether required… |
rippling_list_departments | List departments with name, reference code, parent department and the hierarchy of department ids from the top |
rippling_get_department | Read one department by id: name, reference code, parent and the chain of parent departments |
rippling_list_teams | List teams with their name and parent team, to resolve the teams_id values on workers |
rippling_get_team | Read one team by id: name and parent team (expand parent for its details) |
rippling_list_work_locations | List work locations (offices) with name and full address, to resolve a worker's location |
rippling_get_work_location | Read one work location by id (the location.work_location_id of a worker) |
rippling_list_levels | List job levels with name, parent level, track and rank, to resolve the level_id of workers |
rippling_get_level | Read one job level by id: name, parent level and track (expand track for its name) |
rippling_list_titles | List the job titles defined in the company, with id and name, for consistent title values |
rippling_list_employment_types | List employment types: label, CONTRACTOR or EMPLOYEE, HOURLY or SALARIED, FULL-TIME, PART-TIME or TEMPORARY |
rippling_list_job_codes | List job codes (the codes workers log time against) with their job dimension, department and work location |
rippling_list_supergroups | List supergroups: the attribute-based groups Rippling uses for policies, apps and workflows (name, type, description, whether terminated… |
rippling_get_supergroup | Read one supergroup by id: name, type, description and membership settings |
rippling_list_supergroup_members | List the members of a supergroup: worker id, full name and work email |
rippling_list_leave_types | List leave types (vacation, sick and others): name, type, description, whether paid and whether managed by an external system |
rippling_list_leave_balances | List leave balances per worker and leave type: whether unlimited, balance with and without future requests, in hours and in days |
rippling_get_leave_balance | Read one leave balance by id: worker, leave type and remaining balance in hours and days |
rippling_list_leave_requests | List leave requests with worker, requester, reviewer, status (PENDING, APPROVED, REJECTED, CANCELED), dates, hours, leave type and comments |
rippling_get_leave_request | Read one leave request by id: worker, status, dates, hours, leave type, reviewer and comments |
rippling_list_time_entries | List time entries (worked shifts with job shifts, breaks, duration, status DRAFT, APPROVED, PAID or FINALIZED, pay period and tags) |
rippling_get_time_entry | Read one time entry by id: job shifts, breaks, duration, status, pay period and comments |
rippling_create_department | Create a department, optionally under a parent department and with a reference code |
rippling_update_department | Change a department's name, parent or reference code |
rippling_create_team | Create a team, optionally nested under a parent team; assign workers to it in Rippling |
rippling_update_team | Rename a team or move it under another parent team |
rippling_delete_team | Delete a team; workers lose the team assignment |
rippling_create_work_location | Create a work location (office) with its name and postal address, for assigning workers |
rippling_update_work_location | Rename a work location or change its address |
rippling_create_leave_request | File a leave request for a worker |
rippling_update_leave_request | Approve, reject or cancel a leave request (status), or change its dates and comments |
rippling_create_time_entry | Record worked time for a worker, as job shifts with start and end times or as a duration on a day |
rippling_update_time_entry | Change a time entry: shifts, breaks, duration, comments, tags or status. worker_id is required by the API even when unchanged |
rippling_delete_time_entry | Delete a time entry permanently, e.g. a duplicate |
rippling_create_draft_hires | Create draft hires that then follow Rippling's review and approval policies before anyone is hired |
示例提示
- “最近 30 天谁入职了?列出姓名、职位、部门、经理和入职日期。”
- “列出所有待审批的休假申请,附上员工、日期、小时数和休假类型。”
- “Maria Lopez 还剩多少年假,按天和按小时分别是多少?”
- “批准 Maria Lopez 11 月 10 日至 14 日待审批的休假申请。”
- “每个部门和每个工作地点各有多少在职员工?”
FAQ
Claude Code 和 Claude Desktop 一样工作吗? 是的,两者都指向 https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID。
下一步
这份指南对你有帮助吗?