R Knowledge MCP
A read-only MCP server for source-attributed R package, documentation, compatibility, and workflow knowledge.
Search and inspect CRAN/Bioconductor packages: metadata, dependencies, vignettes, raw Rd function docs, and version checks.
Check R/package compatibility and map R versions to official Bioconductor releases.
Retrieve optional GitHub
AGENTS.mdguidance and Context7 docs with verification and redaction safeguards.Build, improve, validate, and ground R prompts; list prompt templates and resolve package context.
Access official Shiny, Positron, and Quarto documentation and guidance.
Statically analyze supplied Shiny/Quarto/workspace/code files and visualization pipelines (ggplot2, ggraph, tidyr, plotly, knitr) without executing code.
Get combined R workflow and visualization guidance backed by official docs and package knowledge.
All tools are read-only, do not execute R code, and report provenance/verification status where applicable.
Retrieves AGENTS.md files from R packages' metadata-declared GitHub repositories at an explicit ref, pinned to a commit SHA, for package agent guidance.
Provides exact current CRAN Rd documentation for plotly functions and static analysis of documented ggplotly conversion caveats and compatibility.
Retrieves official Positron documentation and guidance, optionally using version, OS, R, and project metadata, with an explicit caveat that current docs are not version-pinned.
Retrieves official Quarto documentation and supports static analysis of Quarto documents and projects, including YAML front matter, R chunk options, formats, citations, and manifest-level bibliography checks without rendering.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@R Knowledge MCP what are the dependencies of ggplot2?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
R Knowledge MCP
Read-only MCP server providing source-attributed information about current R packages in CRAN and Bioconductor. Python 3.11+, official Python MCP SDK (FastMCP), async httpx, safe YAML release discovery. MIT licensed. CRAN/Bioconductor tools work without credentials. Context7 is an optional external integration and requires a server-side API key.
Install and run
git clone https://github.com/menshikovs/R-Knowledge-MCP.git
cd R-Knowledge-MCP
python -m venv .venv
# Linux/macOS:
source .venv/bin/activate
# Windows PowerShell instead: .venv\Scripts\Activate.ps1
python -m pip install -e '.[dev]'
r-knowledge-mcp # stdio (default)
r-knowledge-mcp --transport streamable-http # http://127.0.0.1:8000/mcpThe cran-knowledge-mcp command remains available as a backward-compatible
alias, and the Python import package remains cran_knowledge_mcp.
The project pins mcp>=1.30,<2. SDK 2 is the current stable major and renamed FastMCP to MCPServer; using the maintained 1.x line preserves the requested FastMCP API. See the official SDK documentation and migration guide.
Related MCP server: MCPyDoc
Connect Codex
Use an absolute executable path from the installed virtual environment:
codex mcp add r-knowledge -- /absolute/path/R-Knowledge-MCP/.venv/bin/r-knowledge-mcp
codex mcp listOn Windows the executable is .venv\Scripts\r-knowledge-mcp.exe.
Alternatively put this in ~/.codex/config.toml (use your actual absolute path):
[mcp_servers.r_knowledge]
command = "/absolute/path/R-Knowledge-MCP/.venv/bin/r-knowledge-mcp"For a running local HTTP server:
codex mcp add cran-knowledge-http --url http://127.0.0.1:8000/mcpOther MCP clients can launch the same stdio command or connect to /mcp using
Streamable HTTP. The HTTP server is stateless and uses JSON responses. A ChatGPT
remote connection needs reachable HTTPS hosting; this repository does not deploy
or provide a hosted endpoint. Codex configuration follows
official OpenAI documentation.
Tools and repository selection
The default surface (MCP_TOOLSET=consolidated) has 15 tools. Every parameter has
a schema description, every tool has a title and readOnlyHint=true, and
openWorldHint is false only for tools that never contact an upstream service.
MCP_TOOLSET=legacy restores the original 35 tools unchanged (used by the Codex
benchmark conditions); resources and prompts are registered in both modes.
Tool | Main arguments | Behavior | Replaces (legacy) |
|
| Ranked name/title search; every term must match | same name |
|
| One call returns any of: metadata, dependencies, vignettes, agent_guidance (needs |
|
|
| Raw Rd from the verified current source archive; a different requested version is an explicit error |
|
|
| Compare a client-supplied version with latest CRAN | same name |
|
| CRAN or Bioconductor R and direct dependency constraints; without |
|
|
| BM25 (SQLite FTS5) over the local index of six official R manuals | same name |
|
| One indexed manual section | same name |
|
| Page from the fixed official catalog |
|
|
| Framework scenarios combine docs and package knowledge; visualization scenarios group current Rd references |
|
|
| Bounded static analysis; options that the selected kind cannot use are rejected, not ignored |
|
|
| Workspace metadata; adds Quarto project analysis when |
|
| objective, project type, requirements, stack, target, preferences | Provisional architecture with repository-checked candidates | same name |
|
| Build a prompt, or improve an existing one verbatim |
|
|
| Heuristic structural checklist | same name |
|
| Verified package context for prompt grounding | same name |
Tools that return repository data accept verbosity="compact" (default) or
"full". Compact responses omit repeated static notes (limitations,
verification_notes) and merge nested sources lists into one deduplicated
top-level list; nested results keep their own source_url and checked_at, and
safety fields such as trust and not_executed are always present. Results that
still exceed 60,000 bytes have their longest text fields shortened and list each
cut in output_truncation.
Resources (both tool sets):
URI | Content |
| Fixed Shiny, Positron and Quarto topic catalog |
| Built-in prompt template identifiers |
| Section ids and headings of one indexed R manual |
| One manual section (up to 32,000 characters) |
| First page of raw Rd for one alias from current CRAN |
repository accepts cran, bioconductor, and all. Existing calls default
to CRAN. all keeps separate identities when names overlap, and exposes source
failures in errors with verification_status="partial" or "failed". Absence
from a checked index differs from an unavailable source. Combined search orders
CRAN results then Bioconductor results, using a shared offset/limit (1-20).
Within a repository, results rank exact names first, then name prefixes and name
words, then title words; title terms match whole words or word prefixes, so
api does not match therapy.
See Technology stack recommendations for
the input example, evidence contract, and limits of recommend_technology_stack.
The legacy get_package_info keeps deep archive verification as its default;
get_package defaults to verify_archive=false. Without archive verification:
CRAN reads the official per-package DESCRIPTION and checks its exact package and
version against PACKAGES; if that optional endpoint returns 404, the tool returns
partial index-only metadata and identifies missing fields. Bioconductor returns
metadata available in the selected release's software, annotation, and experiment
PACKAGES index. It does not fetch a current package page or substitute another
release when DESCRIPTION fields are absent.
Inspect metadata_verification_level before relying on a result:
{
"package": "ExamplePkg",
"repository": "cran",
"package_version": "2.10-1",
"bioc_version": null,
"verification_status": "verified",
"metadata_verification_level": "description_version_matched",
"source_url": "https://cran.r-project.org/web/packages/ExamplePkg/DESCRIPTION",
"checked_at": "2026-10-22T00:00:00+00:00"
}index_verified means the record came from the selected repository index;
description_version_matched means a DESCRIPTION identity matched that index;
archive_checksum_verified means the indexed archive checksum was present and
matched. If an archive has no indexed checksum, deep mode still checks its
DESCRIPTION package/version and reports description_version_matched.
verification_status continues to describe whether the returned data was
retrieved and validated; it does not imply the strongest assurance level.
Only get_package_info exposes the lightweight option. get_package_knowledge,
GitHub AGENTS.md, Context7, Prompt Intelligence package grounding, and function
Rd retrieval retain deep verification by default. Dependencies and Bioconductor
compatibility already use the selected indices. See source assurance and
performance measurements.
Prompt Intelligence also registers fourteen standard MCP Prompts, including
r-multi-stage-workflow, which uses the same bounded local composition pipeline
as build_r_prompt but does not query package providers. The other prompts cover package
development, Shiny, Quarto, Bioconductor, ggplot2, debugging, architecture,
CSV import/cleaning/validation, SQL queries/database connections, and CSV-to-SQL.
They are user-selected, deterministic drafts. Connect clients that
support MCP Prompts through their normal prompt discovery UI; MCP tools
remain available for clients that invoke tools directly. These prompts do not
inspect a local workspace or fetch package data, so their supplied context is
explicitly unverified. Tools may fetch bounded package context from the existing
official providers. Set PROMPT_INTELLIGENCE_ENABLED=false to disable the new
tools and prompts without affecting existing knowledge tools.
Prompt construction, examples, configuration, limitations, and evaluation guidance are documented in Prompt Intelligence and the Composable Prompt Engine, which includes a complete generated CSV → PostgreSQL → dplyr → ggplot2 → Quarto prompt. The deterministic design and external optimizer comparison are in Prompt Engineering Research. The complete built-in prompt catalog and each template's package/API candidates are listed in Prompt Templates.
For Bioconductor, bioc_version accepts release (default), stable (alias),
devel, or a specific historical release in the official map. Numbers are
discovered dynamically, then used in pinned URLs. Unavailable history never falls
back to release. Software, annotation and experiment data indices are searched
together; results identify the category. PACKAGES often omits titles, so search
can be name-only; missing titles are not generated.
Example MCP argument objects (not claims about current versions/compatibility):
{"query":"DESeq2","repository":"bioconductor"}
{"query":"ggplot2","repository":"cran"}
{"package":"DESeq2","repository":"bioconductor","bioc_version":"devel"}
{"package":"DESeq2","function":"DESeq","repository":"bioconductor"}
{"r_version":"4.4.2"}
{"package":"DESeq2","r_version":"4.4.2"}
{"package":"DESeq2","repository":"all","bioc_version":"3.20"}get_bioconductor_version returns all mapped releases for an R major/minor and
recommends the latest released one. It never implicitly recommends devel.
Unknown R versions get no invented recommendation. Without an explicit release,
check_package_compatibility chooses a released version mapped to the supplied
R, or returns no_compatible_release.
Verification and compatibility
Every application result includes package, repository, package_version,
bioc_version, source_url, checked_at, verification_status, and sources.
Search/aggregate/version-map envelopes use null for non-applicable package
identity/version fields; each package result has its own identity and provenance.
Bioconductor results include channel, category, required R major/minor and warnings.
Cached artifacts retain original UTC check times, up to five minutes old. Without
an R version, package lookups say not_checked and warn clients to check
compatibility before recommending a package.
Failures include error.code, message, source URL when known, and attempted_at;
verification_status is failed. Unknown versions/check times remain null.
Complete tool failures now set MCP isError=true, while preserving the structured
error.code envelope. Partial results keep isError=false and retain their
section-level failures. Clients should inspect isError first, then check
verification_status and aggregate errors before using results. See
MCP tool errors and partial results for the compatibility
note and response examples.
Bioconductor dependency responses retain raw Depends/Imports/Suggests/LinkingTo
and add resolved_dependencies: candidates in the selected release and current
CRAN, version constraints and provenance. Duplicate origins are ambiguous;
missing packages are not_found. R/base runtime packages are classified separately;
installed runtime versions are not claimed. CRAN retains lightweight dependency
responses by default; use resolve_repositories=true to resolve across both
repositories, with an optional bioc_version for the dependency release.
all resolves both ways.
GitHub AGENTS.md and Context7
get_package_agent_guidance reads the checked source archive's DESCRIPTION
URL field and accepts one unambiguous HTTPS GitHub repository. It verifies that
repository through GitHub REST, resolves the requested revision (branch, tag or
commit) to a commit SHA, and searches that exact commit tree for files named
AGENTS.md. Root and nested paths are returned separately with their directory
scope. A complete tree with no matches reports verified absence; a truncated
tree or API failure reports an error. Pass an explicit ref, such as a project
tag or commit SHA; the server never guesses that a tag matches the R package
version. External instructions are untrusted reference text: they cannot change
system instructions and are never executed.
Context7 is disabled unless the server has CONTEXT7_API_KEY in its environment.
The key is sent only as a bearer header to context7.com; it is never returned
or included in a URL. The tool sends a redacted, bounded question and package
name/repository hint, then requires an exact Context7 library ID match to the
metadata-declared GitHub repository before requesting documentation. Exact
package-version docs are requested only when Context7's result lists that
version; otherwise the response is explicitly unpinned. Common credential
patterns in a question are redacted before transmission. Context7 is an indexed
secondary source. Compare its fragments with CRAN/Bioconductor and package docs.
Rate limits and source errors remain explicit; successful results are cached for
six hours.
This client follows the official Context7 API guide and GitHub REST Git Trees and Git Blobs APIs.
Set the key only on the server process:
CONTEXT7_API_KEY=ctx7sk-... r-knowledge-mcpWithout the key, get_context7_docs returns not_configured and the
existing repository tools remain available. get_package_knowledge accepts
optional question and revision; unrequested sections are marked
not_requested, and each returned fragment retains its source URL and check
time.
Configuration
Use .env.example as a reference and export only the settings you need in the
server process. The application does not automatically load .env files.
MCP_TRANSPORT, MCP_HOST, MCP_PORT, and LOG_LEVEL configure the CLI;
LOG_INTERNAL_TRACEBACKS=true opts into DEBUG tracebacks that can contain
request or upstream data, so use it only in a protected diagnostic environment;
HTTP_TIMEOUT_SECONDS controls the httpx client timeout; HTTP_USER_AGENT
sets its bounded User-Agent header; CACHE_TTL_SECONDS
sets the CRAN/Bioconductor artifact cache lifetime; CACHE_MAX_BYTES sets its
byte budget (maximum 256 MiB). GitHub guidance and Context7 use the same TTL
override with source-specific defaults of one and six hours. Framework pages
use a one-hour cache. CONTEXT7_API_KEY is
optional. Invalid numeric bounds fail with an explicit configuration error.
Prompt settings include PROMPT_INTELLIGENCE_ENABLED,
PROMPT_MAX_INPUT_CHARS, PROMPT_MAX_OUTPUT_CHARS,
PROMPT_MAX_GROUNDING_PACKAGES, PROMPT_DEFAULT_TARGET, and
PROMPT_DEFAULT_DETAIL_LEVEL. See the prompt guide for bounds and meanings.
MCP_TOOLSET selects consolidated (default, 15 tools) or legacy (the original
35). GITHUB_TOKEN is optional; it raises the GitHub API limit for AGENTS.md
lookups from 60 to 5,000 requests per hour, is sent only as an Authorization
header to api.github.com, and never appears in URLs, logs or results.
R_KNOWLEDGE_NATIVE=0 forces the pure-Python implementations even when the
optional Rust extension is installed.
CRAN and Bioconductor indexes are fetched as PACKAGES.gz (falling back to
PACKAGES only when the compressed file is absent). Expired cache entries are
revalidated with If-None-Match/If-Modified-Since; a 304 keeps the parsed
index and records the new checked_at, while any error still fails explicitly
instead of serving stale data.
Compatibility returns metadata_compatible, incompatible or unknown, with a
nullable compatible flag. For CRAN it checks the current record's R (...)
constraint and its direct required dependencies in the CRAN index; dependencies
outside CRAN, or a patch-level bound for an R minor version such as 4.1, make
the result unknown. Only the current CRAN version is checked. For Bioconductor
it checks the official R/release mapping, package R
constraints including patch requirements, and direct required dependency
availability/version/R constraints. It does not solve transitive dependencies,
inspect installed packages, or test platform binaries/system libraries.
Historical releases use current CRAN candidates and warn that these may not match
the historical environment. A successful metadata check is not an installation test.
latest_cran_version is retained only on CRAN results, separately from explicitly
unverified, client-supplied installed_version. Neither provider infers local state.
Official sources and architecture
CRAN uses /src/contrib/PACKAGES, its name/title catalog and exact indexed source
archives. Recommended Path overrides cannot replace canonical root records.
Bioconductor uses:
config.yaml: release/devel and
r_ver_for_bioc_ver, also used by BiocManager. No release number is fixed in code./packages/{version}/bioc/src/contrib/PACKAGES: software./packages/{version}/data/annotation/src/contrib/PACKAGES: annotation data./packages/{version}/data/experiment/src/contrib/PACKAGES: experiment data.Exact indexed source archives for DESCRIPTION and raw
man/*.Rd.Numeric versioned official package pages for vignette links. Page/index disagreement is an error; linked content is not downloaded or executed.
Archive identity and available index MD5 are checked. MD5 detects consistency errors; HTTPS provides transport authenticity. No silent version substitution. Sources reviewed: installation, release/R history, BiocManager mapping, Writing R Extensions. Source formats and live package requests were checked on 2026-10-22.
src/cran_knowledge_mcp/
server.py MCP registration, lifecycle, stdio/HTTP CLI
service.py routing, provenance, combined results, compatibility
releases.py dynamic official Bioconductor/R map
providers/protocol.py PackageRepository extension contract
providers/base.py repository-neutral source/document operations
providers/cran.py CranProvider
providers/bioconductor.py BioconductorProvider, three categories per release
providers/github_agents.py pinned GitHub tree/blob reader for untrusted AGENTS.md
providers/context7.py optional, repository-matched Context7 client
http.py independent fixed-route policies, async HTTP, TTL/LRU
parsers.py bounded data-only DCF, HTML, gzip/tar and Rd aliasesSee architecture details, source authority, security model, and development rules. For framework behavior and limits, see the Shiny, Positron and Quarto integration guide. Visualization, tidyr and reproducible reporting analysis is described in the visualization and reporting guide.
Limits and security
No arbitrary outbound URLs, redirects, environment proxies, archive extraction, R installation, package execution or Rd evaluation. Vignette links are restricted to exact official host/package/release paths. Package text remains untrusted data.
CRAN/Bioconductor artifacts use a configurable 5-minute default TTL and a 64 MiB / 32-artifact LRU budget. Framework pages use a one-hour cache; GitHub AGENTS.md uses a configurable one-hour default, and Context7 uses a configurable six-hour default. Provider caches are bounded and failed requests never replace fresh data. GitHub and Context7 use fixed HTTPS hosts and routes. No caller URL is fetched, redirects are rejected, GitHub tree/blob sizes and per-file/total AGENTS.md bytes are limited, and all external text remains untrusted. At most eight Bioconductor release providers are retained per service. Downloads are serialized per transport. Connect/read timeout is 15 seconds, download deadline 45 seconds, tool deadline 60 seconds. No error cache or stale fallback.
Config: 1 MiB; index/catalog/page: 16 MiB; source download: 25 MiB; expanded tar: 128 MiB; members: 20,000; selected file: 2 MiB; selected files: 16 MiB. Oversized annotation/experiment archives fail explicitly; index-level search/dependencies still work. JSON tool data: 60,000 UTF-8 bytes; raw Rd: 8,000 characters per page; direct dependency resolution: 250 entries.
HTTP defaults to loopback with DNS rebinding protection. A future domain can use
repeatable --allowed-host cran.example.org (include port if needed). Public
hosting needs TLS, ingress rate/body/concurrency limits and an egress policy.
There is no deployment, hosted endpoint, authentication or per-client rate
limiting here. Upstream documentation retains its package license.
Docker
docker build -t r-knowledge-mcp .
docker run --rm -p 127.0.0.1:8000:8000 r-knowledge-mcp
# stdio:
docker run --rm -i r-knowledge-mcp --transport stdioThe container runs as an unprivileged user. Binding 0.0.0.0 inside the container
allows port forwarding; HTTP Host validation remains enabled.
Develop and verify
python -m pip install -e '.[dev]'
ruff check .
ruff format --check .
pytestOptional Rust accelerators live in native/ (PyO3, abi3 wheel for CPython
3.11+). They speed up R source scanning (5–9x on 250 KiB inputs) and package
search ranking (about 12x on 22k packages). The pure-Python code remains the
reference: without the extension everything works identically, and
tests/test_native_parity.py compares both implementations on generated inputs
(it is skipped when the extension is absent).
cd native
maturin build --release -i ../.venv/bin/python # Windows: ..\.venv\Scripts\python.exe
pip install target/wheels/cran_knowledge_native-*.whl
cargo clippy --release -- -D warningsCI tests Python 3.11–3.14 and builds Docker without pushing an image. Unit and
protocol tests mock HTTP and need no CRAN access. See AGENTS.md.
For optional live checks after installation, run python scripts/smoke_cran.py
or python scripts/smoke_repositories.py. The scheduled
python scripts/smoke_upstream.py check covers selected official CRAN,
Bioconductor, Shiny, Positron and Quarto sources. It also runs manually through
the Official upstream smoke checks Actions workflow on the default branch.
Live source checks are separate from ordinary PR CI because upstream outages
are outside the repository's control. Coverage, failure categories and limits
are described in Upstream monitoring.
Known limitations and next stage
Documentation is raw Rd with literal aliases; macros and conditional content are not rendered. Vignettes are verified link listings, not a claim that code was run or every linked document was downloaded. Bioconductor history requires an official mapping and all three retrievable indices. Missing/redirected legacy endpoints fail explicitly. Workflows/books are outside the requested categories.
No local inspection, renv lock restoration, archived CRAN version lookup,
transitive solver, installation test or public hosting is included. Visualization
tools can inspect bounded lock JSON explicitly supplied by the client, without
restoring or verifying an installed library. See the roadmap.
analyze_r_workspace only sees text provided by the client. Shiny and Quarto
analyzers are static checks, not runtime, schema-complete, browser or security
tests.
Optional live checks after installation: python scripts/smoke_cran.py,
python scripts/smoke_repositories.py and python scripts/smoke_upstream.py.
Available Tools
32 toolsanalyze_ggplot_codeCRead-only
Statically inspect ggplot calls and optional schemas; R source is never run.
Inputs: source (required), renv_lock, data_schema, package_versions. Returns: object
analyze_ggplotly_compatibilityBRead-only
Find documented ggplotly caveats; full compatibility is not claimed.
Inputs: source (required), renv_lock, package_versions. Returns: object
analyze_ggraph_codeCRead-only
Statically inspect ggraph calls, separate node/edge mappings, and optional schemas.
Inputs: source (required), renv_lock, edge_schema, node_schema, +1 more. Returns: object
analyze_knitr_chunksCRead-only
Inspect knitr and Quarto chunk options statically; never render or execute chunks.
Inputs: source (required), renv_lock, document_type, package_versions. Returns: object
analyze_quarto_documentCRead-only
Safely parse QMD YAML and fenced R chunk options without rendering or executing code.
Inputs: source (required), r_version, dependencies, project_config, +1 more. Returns: object
analyze_quarto_projectBRead-only
Analyze supplied Quarto YAML and a path-only manifest; no server filesystem access.
Inputs: project_config (required), file_manifest, metadata_config. Returns: object
analyze_r_workspaceBRead-only
Analyze supplied project text; never reads paths or exposes .Renviron values.
Inputs: files (required). Returns: object
analyze_shiny_appBRead-only
Run bounded heuristic checks on supplied Shiny source; never executes it.
Inputs: files, source, shiny_version. Returns: object
analyze_tidyr_pipelineCRead-only
Statically inspect tidyr operations; data-dependent results are not inferred.
Inputs: source (required), renv_lock, input_schema, expected_schema, +1 more. Returns: object
build_r_promptCRead-only
Build one editable R prompt; compose detected workflow stages without executing them.
Inputs: request (required), target, language, packages, +14 more. Returns: object
check_package_compatibilityCRead-only
Check R/release and direct required dependency constraints; installation is untested.
Inputs: package (required), r_version (required), bioc_version. Returns: object
check_package_versionARead-only
Compare a client-supplied R package version to CRAN; never claims local verification.
Inputs: package (required), installed_version (required). Returns: object
explain_shiny_reactivityCRead-only
Explain a Shiny topic with official docs and optional static source checks.
Inputs: topic (required), question (required), source. Returns: object
get_bioconductor_versionARead-only
Map R to official Bioconductor releases; never recommend devel implicitly.
Inputs: r_version (required). Returns: object
get_context7_docsCRead-only
Fetch Context7 docs after matching its ID to verified package metadata.
Inputs: package (required), question (required), repository, bioc_version. Returns: object
get_dependenciesCRead-only
Get direct dependency fields; Bioconductor also resolves CRAN/Bioconductor origins.
Inputs: package (required), repository, bioc_version, resolve_repositories. Returns: object
get_function_docsCRead-only
Get raw Rd by literal alias from the selected repository and release source archive.
Inputs: package (required), function (required), offset, repository, +1 more. Returns: object
get_package_agent_guidanceBRead-only
Read AGENTS.md at a metadata-verified GitHub ref; never execute it.
Inputs: package (required), revision (required), repository, bioc_version. Returns: object
get_package_infoCRead-only
Get metadata with index/DESCRIPTION lookup or deep archive verification.
Inputs: package (required), repository, bioc_version, verify_archive. Returns: object
get_package_knowledgeCRead-only
Combine package metadata with requested AGENTS.md and Context7 material.
Inputs: package (required), question, revision, repository, +1 more. Returns: object
get_package_vignettesBRead-only
List vignette links from the version-checked official page; never execute content.
Inputs: package (required), repository, bioc_version. Returns: object
get_positron_guidanceBRead-only
Retrieve official Positron guidance; this does not control the IDE or an R session.
Inputs: topic (required), r_version, project_type, operating_system, +1 more. Returns: object
get_quarto_docsCRead-only
Retrieve a page from the fixed catalog of official Quarto documentation.
Inputs: topic (required), quarto_version. Returns: object
get_r_prompt_contextCRead-only
Resolve selected R package versions, direct dependencies and source provenance.
Inputs: packages (required), r_version, bioconductor_release. Returns: object
get_r_workflow_guidanceCRead-only
Combine official framework docs with knowledge for explicitly named packages.
Inputs: scenario (required), packages, r_version, bioc_version. Returns: object
get_shiny_docsBRead-only
Get genuine Rd for a Shiny function from a CRAN source archive.
Inputs: package (required), function (required), package_version. Returns: object
get_visualization_docsCRead-only
Get exact current CRAN Rd by alias; historical versions are not substituted.
Inputs: package (required), function (required), package_version. Returns: object
get_visualization_workflowDRead-only
Group current official function references; no code is generated or run.
Inputs: scenario (required), package_versions. Returns: object
improve_r_promptCRead-only
Structure an existing prompt while retaining its complete original text and constraints.
Inputs: existing_prompt (required), target, packages, r_version, +6 more. Returns: object
list_r_prompt_templatesARead-only
List stable built-in R prompt template identifiers and descriptions.
Inputs: no named parameters. Returns: object
search_packagesCRead-only
Search repository names and available title keywords; repository defaults to cran.
Inputs: query (required), limit, offset, repository, +1 more. Returns: object
validate_r_promptCRead-only
Check prompt structure, conflicts, possible secrets and unverified API claims.
Inputs: prompt_text (required). Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
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.
Related MCP Connectors
Real-time package and vulnerability data (Python, npm, Cargo, Go, and more) for AI coding agents.
Versioned documentation registry and semantic search for AI tools and coding assistants.
Search and read Rust documentation for the standard library and any crate on crates.io
@latest documentation and code examples to 9000+ libraries for LLMs and AI code editors in a singl…
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to search documentation of packages and services to find implementation details, examples, and specifications.MIT
- AlicenseNot gradedqualityDmaintenanceProvides real-time access to Python package documentation, source code, and symbol search to prevent AI hallucinations.21 PyPI7MIT
- AlicenseAqualityBmaintenanceEnables Claude to interact with R package development workflows, including CRAN metadata lookup, dependency checking, testing, documentation validation, and release automation.6MIT
- AlicenseNot gradedqualityAmaintenanceEnables Claude Code and other compatible clients to fetch real-time npm package and framework documentation, TypeScript signatures, code examples, version diffs, migration guides, error diagnoses, package comparisons, and dependency scans. It delivers context-aware, intent-formatted results with caching and multi-source integration from npm and GitHub.MIT