R Knowledge MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_HOST | No | Host for the HTTP transport. HTTP defaults to loopback. | |
| MCP_PORT | No | Port for the HTTP transport. The local HTTP endpoint is served at /mcp. | 8000 |
| LOG_LEVEL | No | Logging level for the server. | |
| MCP_TRANSPORT | No | MCP transport to use, e.g. stdio (default) or streamable-http. | stdio |
| CACHE_MAX_BYTES | No | Sets the cache byte budget. Maximum is 256 MiB. | 67108864 |
| HTTP_USER_AGENT | No | Sets the bounded User-Agent header used by the httpx client. | |
| CONTEXT7_API_KEY | No | Optional Context7 API key. Context7 is disabled unless this is set; the key is sent only as a bearer header to context7.com. Without it, get_context7_docs returns not_configured. | |
| CACHE_TTL_SECONDS | No | Sets the CRAN/Bioconductor artifact cache lifetime. Default is 5 minutes. | 300 |
| HTTP_TIMEOUT_SECONDS | No | Controls the httpx client timeout. | |
| PROMPT_DEFAULT_TARGET | No | Prompt Intelligence default target setting. | |
| PROMPT_MAX_INPUT_CHARS | No | Prompt Intelligence maximum input characters setting. | |
| LOG_INTERNAL_TRACEBACKS | No | Set to true to opt into DEBUG tracebacks that can contain request or upstream data; use only in a protected diagnostic environment. | false |
| PROMPT_MAX_OUTPUT_CHARS | No | Prompt Intelligence maximum output characters setting. | |
| PROMPT_DEFAULT_DETAIL_LEVEL | No | Prompt Intelligence default detail level setting. | |
| PROMPT_INTELLIGENCE_ENABLED | No | Set to false to disable Prompt Intelligence tools and prompts without affecting existing knowledge tools. | true |
| PROMPT_MAX_GROUNDING_PACKAGES | No | Prompt Intelligence maximum grounding packages setting. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| build_r_promptC | Build one editable R prompt; compose detected workflow stages without executing them. |
| improve_r_promptC | Structure an existing prompt while retaining its complete original text and constraints. |
| validate_r_promptC | Check prompt structure, conflicts, possible secrets and unverified API claims. |
| list_r_prompt_templatesA | List stable built-in R prompt template identifiers and descriptions. |
| get_r_prompt_contextC | Resolve selected R package versions, direct dependencies and source provenance. |
| search_packagesC | Search repository names and available title keywords; repository defaults to cran. |
| get_package_infoC | Get metadata with index/DESCRIPTION lookup or deep archive verification. |
| get_dependenciesC | Get direct dependency fields; Bioconductor also resolves CRAN/Bioconductor origins. |
| get_function_docsC | Get raw Rd by literal alias from the selected repository and release source archive. |
| check_package_versionA | Compare a client-supplied R package version to CRAN; never claims local verification. |
| get_bioconductor_versionA | Map R to official Bioconductor releases; never recommend devel implicitly. |
| check_package_compatibilityC | Check R/release and direct required dependency constraints; installation is untested. |
| get_package_vignettesB | List vignette links from the version-checked official page; never execute content. |
| get_package_agent_guidanceB | Read AGENTS.md at a metadata-verified GitHub ref; never execute it. |
| get_context7_docsC | Fetch Context7 docs after matching its ID to verified package metadata. |
| get_package_knowledgeC | Combine package metadata with requested AGENTS.md and Context7 material. |
| get_shiny_docsB | Get genuine Rd for a Shiny function from a CRAN source archive. |
| explain_shiny_reactivityC | Explain a Shiny topic with official docs and optional static source checks. |
| analyze_shiny_appB | Run bounded heuristic checks on supplied Shiny source; never executes it. |
| get_positron_guidanceB | Retrieve official Positron guidance; this does not control the IDE or an R session. |
| analyze_r_workspaceB | Analyze supplied project text; never reads paths or exposes .Renviron values. |
| get_quarto_docsC | Retrieve a page from the fixed catalog of official Quarto documentation. |
| analyze_quarto_documentC | Safely parse QMD YAML and fenced R chunk options without rendering or executing code. |
| analyze_quarto_projectB | Analyze supplied Quarto YAML and a path-only manifest; no server filesystem access. |
| get_r_workflow_guidanceC | Combine official framework docs with knowledge for explicitly named packages. |
| get_visualization_docsC | Get exact current CRAN Rd by alias; historical versions are not substituted. |
| get_visualization_workflowD | Group current official function references; no code is generated or run. |
| analyze_ggplot_codeC | Statically inspect ggplot calls and optional schemas; R source is never run. |
| analyze_ggraph_codeC | Statically inspect ggraph calls, separate node/edge mappings, and optional schemas. |
| analyze_tidyr_pipelineC | Statically inspect tidyr operations; data-dependent results are not inferred. |
| analyze_ggplotly_compatibilityB | Find documented ggplotly caveats; full compatibility is not claimed. |
| analyze_knitr_chunksC | Inspect knitr and Quarto chunk options statically; never render or execute chunks. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| r-package-development | Create or extend a maintainable R package. |
| shiny-app-development | Design a Shiny application with clear reactive boundaries. |
| quarto-report | Create a reproducible Quarto document or project. |
| bioconductor-analysis | Plan an R and Bioconductor release-aware analysis. |
| ggplot2-visualization | Develop a ggplot2 visualization from explicit data and aesthetic requirements. |
| r-debugging | Reproduce and diagnose an R failure before changing code. |
| r-architecture-audit | Review an R project from inspected evidence and report risks. |
| csv-import | Inspect and import delimited data with explicit parsing and memory requirements. |
| csv-cleaning | Clean delimited data without assuming column names, types, missingness or duplicate rules. |
| csv-validation | Define measurable data-quality checks and report failures without silently changing data. |
| sql-query | Build a dialect-aware, parameterized query against an explicitly described schema. |
| sql-database | Connect to a named database through DBI with explicit lifecycle and transaction safeguards. |
| csv-sql-pipeline | Plan a validated, restartable delimited-file load with type mapping and transaction-aware checks. |
| r-multi-stage-workflow | Compose one bounded multi-stage R development prompt using the shared task, template, planner and composer pipeline. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 32 tools
The set has many narrowly targeted but similar-sounding tools, such as get_function_docs/get_shiny_docs/get_visualization_docs/get_quarto_docs and the several analyze_* static inspectors. Descriptions differentiate them, but overlapping doc-fetching paths like get_package_knowledge vs get_context7_docs vs get_package_agent_guidance create real misselection risk.
All tool names use snake_case with a consistent verb_noun pattern (get_, check_, analyze_, build_, validate_, list_). The qualifiers are descriptive and predictable despite the length of some names.
32 tools is above the rubric's heavy threshold, with several closely related variants for docs, static analysis, and prompt operations. The broad R-knowledge scope partly justifies it, but the surface feels segmented rather than tightly scoped.
Coverage spans package discovery, metadata, dependencies, version/compatibility checks, docs, agent guidance, Shiny/Quarto/visualization workflows, and prompt lifecycle. Minor gaps remain, such as discovering function aliases or listing package functions, but core knowledge workflows are covered.