No user-submitted related servers found.
Scored across 14 tools
Most tools map to a distinct resource+action: operations (list/describe), artifacts (import/list/inspect/compare/export), datasets (register/get), workflows (validate/run/visualize/export) and runs (get_run). The only soft overlaps are the four *_workflow tools and three *_artifact tools, but their descriptions (static check vs. execute vs. draw vs. compile to cluster jobs; inspect vs. compare vs. write to disk) clearly separate them. get_dataset doubling as 'list datasets' when no id is given is a minor collision with list_artifacts.
Every tool follows a single snake_case verb_noun convention (list_operations, register_dataset, run_workflow, export_artifact). Verbs are reused consistently for the same intent (list_, get_, export_, *_workflow), making the surface predictable.
14 tools sits squarely in the well-scoped range and each one corresponds to a real lifecycle stage (discovery, import/store, dataset registration, workflow validate/run/visualize/export, provenance, artifact export). No obviously redundant or filler tools.
The core pipeline is covered end to end: discover operations, import data, register datasets, validate/run/visualize/export workflows, inspect provenance, and export results for R/statsmodels. Gaps are minor: no delete/prune for artifacts or datasets, no explicit in-memory artifact data read (only summary via inspect or disk write via export), and no workflow listing tool.