UiPath OnPremise MCP Server with multiple endpoints for managing UiPath Orchestrator resources including jobs, queues, schedules, and folders
This server has severe definition quality gaps across all three tools. While tool names follow verb_noun conventions (google_search, uipath_get_jobs_stats, uipath_get_queues_table), the implementations have critical issues: (1) schemas are visible but two tools lack parameter descriptions entirely; (2) descriptions exist but are extremely brief and lack context on WHEN to use tools and WHAT they return; (3) no output schemas documented; (4) error handling guidance absent; (5) the google_search tool appears to be a sample/placeholder with no actual implementation visible. The UiPath tools have slightly better structure (parameters with defaults, time_frame_minutes/page_no/page_size documented) but still lack output documentation and recovery guidance. Code review shows tools are registered in setup scripts (reset_and_add_uipath_tools.py) but actual implementation details are not visible in the provided source.
Google을 통한 검색
Get job statistics from UiPath Orchestrator showing counts by status (Successful, Faulted, Stopped, Running, Pending, etc.)
Get a paginated table of queues with statistics including item counts, SLA status, average handling time, and estimated completion time
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract structured data without knowing what fields will be returned.
Tool descriptions are extremely brief (15-90 chars) and lack context. They do not explain WHEN to use each tool, WHAT data it returns, or HOW it differs from similar tools. Baseline for A+ tools is 194 chars with full context.
Parameter descriptions missing for google_search 'q' parameter. Even though the parameter name is clear, LLMs need explicit guidance on format, length, and examples.
google_search tool has no error handling guidance. If the search fails (API unavailable, quota exceeded, invalid query), the LLM has no recovery path. No pattern:recovery-guide implemented.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 46 | 2026-07-28+ | v2 |
| 2026-03-09 | D | 50 | - | v1 |
uipath_get_jobs_stats has empty required array but description mentions multiple status types (Successful, Faulted, Stopped, Running, Pending). No parameters to filter by status, date range, or other criteria. Tool is overly rigid.
uipath_get_queues_table has pagination support (page_no, page_size, default 1440 minutes) but no documented return structure. LLM cannot know if result includes total_count, next_cursor, or per-item fields needed for chaining.
google_search tool location points to backend/scripts/add_sample_builtin_tool.py but no actual implementation visible. This appears to be a sample/placeholder tool, not a production-ready integration. Actual function implementation (google_search.google_search) not visible in source.