SAS MCP Server
OfficialRelated Servers
Alternatives to SAS MCP Server
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityDmaintenanceExecutes Python code with safety constraints and manages Python packages through the Model Context Protocol.7 npm3MIT
- AlicenseNot gradedqualityDmaintenanceEnables secure cloud-based execution of code across 14+ programming languages within a sandboxed environment. It supports file management, standard input/output handling, and automatic generation of visual artifacts like plots and charts.MIT
- AlicenseNot gradedqualityAmaintenanceEnables secure remote command execution and bidirectional file transfers on SSH servers through the Model Context Protocol. It features robust security controls including command whitelisting, credential isolation, and support for multiple SSH connection profiles.4,268 npm871ISC
- AlicenseNot gradedqualityCmaintenanceEnables AI to execute RPA applications and workflows through the Model Context Protocol, supporting both local and cloud modes.5 npm1MIT
- AlicenseNot gradedqualityBmaintenanceEnables governed, audit-traced AI access to a curated knowledge base through the Model Context Protocol, with OAuth 2.1 authentication and policy enforcement for secure, compliant queries.Apache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables users to execute notebook cells, manage environments, and validate workbooks via MCP, facilitating reproducible machine-learning answers.MIT
TDQS
Scored across 92 tools
Each tool targets a distinct resource and action, with near-zero ambiguity even in dense areas. Overlapping data-access tools (query_data, get_castable_data, get_compute_table_data, execute_sas_code) are precisely differentiated and even cross-reference each other in descriptions. Catalog search vs glossary search, and upload_data vs upload_inline_data vs upload_file, all have clear boundaries.
The dominant verb_noun pattern (list_*, get_*, create_*, update_*, delete_*, lock_*, publish_*) is followed consistently within each domain, and CRUD families are easy to predict. The only deviation is the catalog_* group, which inverts the convention by placing the domain prefix first (catalog_search, catalog_run_agent) rather than as the object noun.
At 92 tools, the surface is far beyond the ideal range and well past the 25+ threshold for heaviness; it would take an agent many calls just to enumerate candidates. That said, the count is defensible: the server spans roughly a dozen genuinely distinct SAS Viya domains (rules, decision flows, CAS, compute, reports, catalog, glossary, ML), each of which is reasonably scoped on its own.
Every major domain has a closed lifecycle: full CRUD plus revisions and locking for business rules and decision flows, complete glossary management including term types and assignments, and report authoring with copy/delete/export. Minor gaps exist — CAS table alteration/removal is absent and MAS module management is read-only — but finding a table, scoring it, or tracing a governed term all have no dead ends.