Automate BambooHR Employee Management

This is a skill for automating BambooHR HR operations through Composio's BambooHR toolkit, accessed via Rube MCP (https://rube.app/mcp, added as an MCP server with no API keys needed on the client side - the toolkit itself still requires an active BambooHR connection authenticated through RUBE_MANAGE_CONNECTIONS). It covers five core workflows: listing and searching employees (BAMBOOHR_GET_ALL_EMPLOYEES for the directory, BAMBOOHR_GET_EMPLOYEE for full details by numeric ID), tracking employee changes for sync or auditing (BAMBOOHR_EMPLOYEE_GET_CHANGED since an ISO 8601 timestamp, returning changed IDs rather than full records), managing time off (listing types, checking balances, creating and approving or denying requests), updating employee profile fields, and managing dependents and benefit coverages. A representative incremental-sync pattern chains these calls together:

1. Store last_sync_timestamp
2. Call BAMBOOHR_EMPLOYEE_GET_CHANGED with since=last_sync_timestamp
3. For each changed employee ID, call BAMBOOHR_GET_EMPLOYEE
4. Process updates in external system
5. Update last_sync_timestamp

Use this skill when automating BambooHR operations - employee lookups, change-tracking for external sync, time-off requests and approvals, profile updates, or dependent and benefit queries - through Rube MCP once the BambooHR connection is confirmed ACTIVE. Always call RUBE_SEARCH_TOOLS first to get current tool schemas rather than assuming parameter names, since the toolkit's own documentation warns fields and schemas can vary, for example benefit-coverage parameters depend on the account's schema. It is not the right tool if the BambooHR connection isn't active yet - the setup flow (verify RUBE_SEARCH_TOOLS, call RUBE_MANAGE_CONNECTIONS for the bamboohr toolkit, complete the returned auth link if not ACTIVE) must complete before any workflow runs.

Input is employee IDs (always numeric), date ranges in YYYY-MM-DD format for time-off, and ISO 8601 timestamps for change detection; output is employee directory and detail records, time-off balances and request statuses, updated employee fields (audited in BambooHR's own change history), and dependent or benefit records. Known data-shape pitfalls are called out explicitly: response data may be nested under a data key, empty fields may be omitted or returned as null, and GET_EMPLOYEE_CHANGED returns only IDs - a separate GET_EMPLOYEE call per ID is required to get the actual changed data. Time-off request creation does not auto-approve the request; a separate update call with manager or admin permissions is needed.

Runs through Rube MCP (https://rube.app/mcp) and Composio's BambooHR toolkit, gated by RUBE_MANAGE_CONNECTIONS for authentication and rate-limited per BambooHR API key - bulk operations should be throttled, and GET_ALL_EMPLOYEES is called out as more efficient than many individual GET_EMPLOYEE calls.

Teams automating BambooHR workflows - HR system syncs, time-off management, employee data updates - through an MCP-connected agent, who need the specific tool sequences, parameter formats, and known pitfalls such as numeric IDs, inconsistent date formats, and PII handling, rather than guessing at the API surface.