Skip to content
Back to search

wolfram-agent-skills

SkillGitHubIncomplete

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/WolframResearch/AgentTools with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/WolframResearch/AgentTools'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/wolframresearch--agenttools/badge.svg)](/listings/wolframresearch--agenttools)

HTML

<a href="/listings/wolframresearch--agenttools"><img src="/api/listings/wolframresearch--agenttools/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Install the Paclet

PacletInstall["Wolfram/AgentTools"]

Load the Package

Needs["Wolfram`AgentTools`"]

Docker Image

A prebuilt Docker image is available as an alternative to a local Wolfram Engine installation. See [Docker documentation](docs/docker.md) for setup and MCP client configuration examples.

docker pull ghcr.io/wolframresearch/mcpserver:latest

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 34 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name":"wolfram-agent-skills",
  3. "owner":{
  4. "name":"Richard Hennigan",
  5. "email":"richardh@wolfram.com"
  6. },
  7. "metadata":{
  8. "description":"A collection of Wolfram agent skills",
  9. "version":"2.2.18"
  10. },
  11. "plugins":[
  12. {
  13. "name":"wolfram-language-development",
  14. "description":"A full Wolfram Language development environment with code evaluation, documentation search, symbol inspection, static analysis, test execution, and debugging.",
  15. "source":"./Assets/AgentSkills",
  16. "strict":false,
  17. "skills":[
  18. "./wolfram-language",
  19. "./wolfram-debugging",
  20. "./wolfram-notebooks",
  21. "./wolfram-paclets"
  22. ]
  23. },
  24. {
  25. "name":"wolfram-alpha",
  26. "description":"Wolfram|Alpha queries and context retrieval.",
  27. "source":"./Assets/AgentSkills",
  28. "strict":false,
  29. "skills":[
  30. "./wolfram-alpha"
  31. ]
  32. }
  33. ]
  34. }
README.md · README · 368 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # [Wolfram/AgentTools](https://paclets.com/Wolfram/AgentTools)
  2. [![GitHub license](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
  3. [![Wolfram Version](https://img.shields.io/badge/Wolfram-15.0%2B-red.svg)](https://www.wolfram.com/language/)
  4. A Wolfram Language toolkit for integrating with AI agents and LLMs — providing [MCP](https://modelcontextprotocol.io) servers, agent skills, and other standard interfaces that give AI systems access to Wolfram's computational capabilities.
  5. ## Table of Contents
  6. - [Features](#features)
  7. - [Requirements](#requirements)
  8. - [Installation](#installation)
  9. - [Quick Start](#quick-start)
  10. - [Predefined Servers](#predefined-servers)
  11. - [Supported Clients](#supported-clients)
  12. - [Available Tools](#available-tools)
  13. - [Creating Custom Servers](#creating-custom-servers)
  14. - [API Reference](#api-reference)
  15. - [Development](#development)
  16. - [License](#license)
  17. ## Features
  18. - **Predefined servers** for common use cases (general computation, Wolfram\|Alpha queries, development)
  19. - **Semantic search** across Wolfram documentation and Wolfram\|Alpha results
  20. - **Code evaluation** with Wolfram Language directly in AI conversations
  21. - **Notebook support** for reading and writing Wolfram notebooks
  22. - **Custom servers** with tailored tools for specific needs
  23. - **Wide client support** including Claude Desktop, Cursor, VS Code, and more
  24. - **MCP prompts** for enhanced context and workflows
  25. - **MCP Apps** for interactive UI resources in supported clients (e.g., embedded notebook viewers, Wolfram\|Alpha result displays)
  26. - **MCP Roots** support so the server, evaluator, and external tools (`TestReport` and similar) resolve relative paths against the client's project directory
  27. - **Agent Skills** for distributing Wolfram tools as portable skills to AI coding agents (Claude Code, Cursor, Gemini CLI, VS Code, and [more](https://agentskills.io/)); the built-in skills (wolfram-language, wolfram-alpha, wolfram-notebooks, wolfram-paclets, wolfram-debugging) ship with the paclet, and `DeployAgentTools` installs them together with the matching MCP server
  28. - **Paclet extensions** allowing third-party paclets to contribute MCP tools, prompts, and servers via the `"AgentTools"` extension
  29. - **Docker image** for running the MCP server in a container without a local Wolfram Engine installation
  30. ## Requirements
  31. - Wolfram Language 15.0 or higher
  32. - An MCP-compatible client application (see [Supported Clients](#supported-clients))
  33. - Optional: [LLMKit subscription](https://www.wolfram.com/notebook-assistant-llm-kit) for enhanced semantic search capabilities
  34. ## Installation
  35. ### Install the Paclet
  36. ```wl
  37. PacletInstall["Wolfram/AgentTools"]
  38. ```
  39. ### Load the Package
  40. ```wl
  41. Needs["Wolfram`AgentTools`"]
  42. ```
  43. ### Docker Image
  44. A prebuilt Docker image is available as an alternative to a local Wolfram Engine installation. See [Docker documentation](docs/docker.md) for setup and MCP client configuration examples.
  45. ```bash
  46. docker pull ghcr.io/wolframresearch/mcpserver:latest
  47. ```
  48. ## Quick Start
  49. Install a Wolfram MCP server for Claude Desktop:
  50. ```wl
  51. InstallMCPServer["ClaudeDesktop"]
  52. (* Out: Success["InstallMCPServer", <|...|>] *)
  53. ```
  54. When no server is specified, each named client gets a sensible default — chat clients (Claude Desktop, Goose) get the general-purpose `"Wolfram"` server, while coding clients (Claude Code, Cursor, VS Code, etc.) get `"WolframLanguage"`. For `File[...]` targets, the same per-client default applies when the path or content identifies a known client (or `"ApplicationName"` is supplied); otherwise it falls back to `"Wolfram"`. Pass an explicit server name as the second argument to override.
  55. After restarting Claude Desktop, it will have access to Wolfram knowledge and tools:
  56. ![Claude Desktop Screenshot](.github/images/sk6raevruc0q.png)
  57. To install a specific server type:
  58. ```wl
  59. InstallMCPServer["ClaudeDesktop", "WolframLanguage"]
  60. ```
  61. To uninstall:
  62. ```wl
  63. UninstallMCPServer["ClaudeDesktop"] (* Remove all servers *)
  64. UninstallMCPServer["ClaudeDesktop", "Wolfram"] (* Remove specific server *)
  65. ```
  66. ## Predefined Servers
  67. AgentTools includes four predefined server configurations, each optimized for different use cases:
  68. | Server | Best For | Tools |
  69. |--------|----------|-------|
  70. | **Wolfram** (default) | General-purpose use combining computational power with natural language | `WolframContext`, `WolframLanguageEvaluator`, `WolframAlpha` |
  71. | **WolframAlpha** | Natural language queries without code execution | `WolframAlphaContext`*, `WolframAlpha` |
  72. | **WolframLanguage** | Wolfram Language development and learning | `WolframLanguageContext`, `WolframLanguageEvaluator`, `ReadNotebook`, `WriteNotebook`, `SymbolDefinition`, `CodeInspector`, `TestReport` |
  73. | **WolframPacletDevelopment** | Developing and maintaining Wolfram paclets | All WolframLanguage tools plus documentation tools (`CreateSymbolDoc`, `EditSymbolDoc`, `EditSymbolDocExamples`) and paclet release tools (`CheckPaclet`, `BuildPaclet`, `SubmitPaclet`) |
  74. *\*Requires [LLMKit subscription](https://www.wolfram.com/notebook-assistant-llm-kit)*
  75. Install a specific server:
  76. ```wl
  77. InstallMCPServer["ClaudeDesktop", "WolframLanguage"]
  78. ```
  79. See [docs/servers.md](docs/servers.md) for detailed information about each server and guidance on choosing the right one.
  80. ## Supported Clients
  81. AgentTools can be installed into the following MCP client applications:
  82. | Client | Name | Project Support |
  83. |--------|------|-----------------|
  84. | [Amazon Q Developer](https://aws.amazon.com/q/developer/) | `"AmazonQ"` | Yes |
  85. | [Augment Code](https://www.augmentcode.com) | `"AugmentCode"` | No |
  86. | [Augment Code IDE](https://marketplace.visualstudio.com/items?itemName=augment.vscode-augment) | `"AugmentCodeIDE"` | No |
  87. | [Claude Code](https://code.claude.com) | `"ClaudeCode"` | Yes |
  88. | [Claude Desktop](https://claude.ai/download) | `"ClaudeDesktop"` | No |
  89. | [Cline](https://cline.bot) | `"Cline"` | No |
  90. | [Continue](https://www.continue.dev/) (VS Code + JetBrains + CLI) | `"Continue"` | Yes |
  91. | [Copilot CLI](https://github.com/features/copilot/cli) | `"CopilotCLI"` | No |
  92. | [Cursor](https://www.cursor.com) | `"Cursor"` | No |
  93. | [Gemini CLI](https://github.com/google-gemini/gemini-cli) | `"GeminiCLI"` | No |
  94. | [Goose](https://block.github.io/goose/) | `"Goose"` | No |
  95. | [Google Antigravity](https://antigravity.google) | `"Antigravity"` | No |
  96. | [Junie](https://www.jetbrains.com/junie/) (JetBrains IDE plugin + CLI) | `"Junie"` | Yes |
  97. | [Kimi Code](https://github.com/MoonshotAI/kimi-cli) | `"KimiCode"` | No |
  98. | [LM Studio](https://lmstudio.ai) | `"LMStudio"` | No |
  99. | [OpenAI Codex](https://openai.com/codex) | `"Codex"` | Yes |
  100. | [OpenCode](https://opencode.ai) | `"OpenCode"` | Yes |
  101. | [Qwen Code](https://github.com/QwenLM/qwen-code) | `"QwenCode"` | Yes |
  102. | [Visual Studio Code](https://code.visualstudio.com) | `"VisualStudioCode"` | Yes |
  103. | [Windsurf](https://codeium.com/windsurf) | `"Windsurf"` | No |
  104. | [Zed](https://zed.dev) | `"Zed"` | Yes |
  105. ### Project-Level Installation
  106. Clients with project support can have servers installed for specific projects:
  107. ```wl
  108. InstallMCPServer[{"ClaudeCode", "/path/to/project"}, "WolframLanguage"]
  109. ```
  110. ### Claude Desktop
  111. Claude Desktop offers an excellent integration experience with AgentTools, providing seamless access to Wolfram Language's computational capabilities.
  112. ### Cursor
  113. Install an MCP server for use in Cursor:
  114. ```wl
  115. InstallMCPServer["Cursor"]
  116. (* Out: Success["InstallMCPServer", <|...|>] *)
  117. ```
  118. Check the MCP tab in Cursor settings to verify the server connection:
  119. ![Cursor MCP Settings Screenshot](.github/images/nldzo3f42xid.png)
  120. Your Wolfram tools will now be available in Cursor agent chat:
  121. ![Cursor MCP Chat Screenshot](.github/images/o6ltldxumzkx.png)
  122. ### Other Clients
  123. AgentTools works with any stdio-based MCP client. See [docs/mcp-clients.md](docs/mcp-clients.md) for manual configuration instructions.
  124. ## Available Tools
  125. AgentTools provides a variety of tools organized by category:
  126. ### Context Tools (Semantic Search)
  127. Search Wolfram resources using semantic similarity:
  128. - **WolframContext** - Combines the functionality of `WolframLanguageContext` and `WolframAlphaContext` in a single tool
  129. - **WolframAlphaContext** - Semantic search of Wolfram\|Alpha results (requires [LLMKit](https://www.wolfram.com/notebook-assistant-llm-kit))
  130. - **WolframLanguageContext** - Semantic search of Wolfram Language documentation and other resources
  131. > **Note:** Since `WolframContext` combines the other two, a server should only include one of these three tools. Without LLMKit, `WolframContext` is effectively the same as `WolframLanguageContext` since the Wolfram\|Alpha semantic search functionality is disabled.
  132. Documentation search includes the [Function Repository](https://resources.wolframcloud.com/FunctionRepository), [Data Repository](https://datarepository.wolframcloud.com), [Neural Net Repository](https://resources.wolframcloud.com/NeuralNetRepository), [Paclet Repository](https://resources.wolframcloud.com/PacletRepository), and more.
  133. While only `WolframAlphaContext` *requires* an [LLMKit subscription](https://www.wolfram.com/notebook-assistant-llm-kit), having LLMKit greatly improves search results for all context tools by enabling reranking and filtering.
  134. ### Code Execution Tools
  135. - **WolframLanguageEvaluator** - Execute Wolfram Language code with time constraints
  136. - **WolframAlpha** - Natural language queries to Wolfram\|Alpha
  137. - **SymbolDefinition** - Retrieve symbol definitions in readable markdown format
  138. ### Notebook Tools
  139. - **ReadNotebook** - Read Wolfram notebooks (.nb) as markdown text
  140. - **WriteNotebook** - Convert markdown to Wolfram notebooks
  141. ### Testing Tools
  142. - **TestReport** - Run Wolfram Language test files (.wlt) and return reports
  143. ### Code Analysis Tools
  144. - **CodeInspector** - Inspect Wolfram Language code for issues and return formatted reports
  145. ### Documentation Tools (Paclet Development)
  146. - **CreateSymbolDoc** - Create new symbol documentation pages
  147. - **EditSymbolDoc** - Edit existing symbol documentation pages
  148. - **EditSymbolDocExamples** - Edit example sections of documentation
  149. ### Paclet Release Tools (Paclet Development)
  150. - **CheckPaclet** - Check a paclet for issues before building or submission
  151. - **BuildPaclet** - Build a `.paclet` archive for distribution
  152. - **SubmitPaclet** - Submit a paclet to the Wolfram Language Paclet Repository
  153. ### MCP Apps Tools
  154. - **NotebookViewer** - Embed interactive Wolfram Cloud notebooks inline
  155. - **MCPAppsTest** - Diagnostic tool for testing the MCP Apps pipeline
  156. > **Note:** MCP Apps tools require a client that supports the `io.modelcontextprotocol/ui` extension. When UI support is available, the `WolframAlpha` and `WolframLanguageEvaluator` tools are automatically enhanced with interactive notebook viewers. See [docs/mcp-apps.md](docs/mcp-apps.md) for details.
  157. ### License Tools
  158. - **WolframLicenseStatus** - Report the status of the standalone MCP server's time-limited license (added automatically to servers running under one)
  159. See [docs/tools.md](docs/tools.md) for detailed information about each tool.
  160. ## Creating Custom Servers
  161. Create custom MCP servers with your own tools using [LLMConfiguration](https://reference.wolfram.com/language/ref/LLMConfiguration.html):
  162. ```wl
  163. config = LLMConfiguration[<|
  164. "Tools" -> {LLMTool["PrimeFinder", {"n" -> "Integer"}, Prime[#n]&]}
  165. |>];
  166. server = CreateMCPServer["My MCP Server", config]
  167. (* Out: MCPServerObject[...] *)
  168. ```
  169. Install for use in Claude Desktop:
  170. ```wl
  171. InstallMCPServer["ClaudeDesktop", server]
  172. (* Out: Success["InstallMCPServer", <|...|>] *)
  173. ```
  174. After restarting Claude Desktop, your custom tools will be available:
  175. ![Claude Desktop Screenshot](.github/images/1j9zrhp9b1y8.png)
  176. You can also mix predefined tools with custom tools:
  177. ```wl
  178. CreateMCPServer["My MCP Server", <|
  179. "Tools" -> {
  180. "WolframLanguageEvaluator", (* Predefined tool *)
  181. "WolframAlpha", (* Predefined tool *)
  182. LLMTool["MyCustomTool", ...] (* Custom tool *)
  183. }
  184. |>];
  185. ```
  186. ## API Reference
  187. ### Core Functions
  188. | Function | Description |
  189. |----------|-------------|
  190. | `CreateMCPServer[name, config]` | Create a custom MCP server |
  191. | `InstallMCPServer[client]` | Install the default server for a client |
  192. | `InstallMCPServer[client, server]` | Install a specific server for a client |
  193. | `UninstallMCPServer[client]` | Remove all servers from a client |
  194. | `UninstallMCPServer[client, name]` | Remove a specific server from a client |
  195. | `DetectedMCPClients[]` | Subset of `$SupportedClients` whose MCP config file exists on this machine |
  196. | `DeployAgentTools[target]` | Deploy tools to a client with tracked deployment management |
  197. | `DeployAgentTools[target, tools]` | Deploy a bundle of MCP servers and agent skills (an `AgentToolsObject` or its name) to a client (see [docs/agent-tools-objects.md](docs/agent-tools-objects.md)) |
  198. | `InstallAgentSkills[client, skills]` | Copy agent skills into a client's skills directory (untracked) |
  199. | `UninstallAgentSkills[client, names]` | Remove agent skills from a client's skills directory (untracked) |
  200. | `DeployAgentTools[All]` | Deploy tools to every client in `$SupportedClients` (see [docs/deploy-agent-tools.md](docs/deploy-agent-tools.md#deploying-to-all-clients)) |
  201. | `ValidateAgentToolsPacletExtension[paclet]` | Validate an `"AgentTools"` paclet extension |
  202. | `DeployedAgentTools[]` | List all tracked deployments |
  203. | `CreatePreferencesContent[]` | Build the preferences panel UI for managing deployed Wolfram toolsets (see [docs/preferences-content.md](docs/preferences-content.md)) |
  204. ### Deployment Objects
  205. | Symbol | Description |
  206. |--------|-------------|
  207. | `AgentToolsDeployment[...]` | Data structure representing a tracked tool deployment |
  208. | `DeployedAgentTools[]` | List all deployments |
  209. | `DeployedAgentTools[client]` | List deployments for a specific client |
  210. ### Agent Tools Bundles
  211. | Symbol | Description |
  212. |--------|-------------|
  213. | `AgentToolsObject[...]` | A bundle of MCP servers and agent skills (built-in, paclet-defined, or ad hoc) |
  214. | `AgentToolsObjects[]` | List the bundles of installed paclets |
  215. | `$DefaultAgentTools` | The built-in bundles, one per built-in MCP server, each with its built-in agent skills |
  216. ### Server Objects
  217. | Symbol | Description |
  218. |--------|-------------|
  219. | `MCPServerObject[...]` | Data structure representing an MCP server |
  220. | `MCPServerObjectQ[expr]` | Test if an expression is a valid server object |
  221. | `MCPServerObjects[]` | List all created server objects (includes paclet-backed servers) |
  222. ### Predefined Resources
  223. | Symbol | Description |
  224. |--------|-------------|
  225. | `$DefaultMCPServers` | Association of predefined server configurations |
  226. | `$DefaultMCPTools` | Association of available tool definitions |
  227. | `$DefaultMCPToolOptions` | Association of default tool option values |
  228. | `$DefaultMCPPrompts` | Association of available prompt definitions |
  229. ### Options
  230. #### `CreateMCPServer`
  231. | Option | Default | Description |
  232. |--------|---------|-------------|
  233. | `OverwriteTarget` | `False` | Overwrite an existing server with the same name |
  234. | `IncludeDefinitions` | `True` | Include function definitions in the serialized server |
  235. | `Initialization` | `None` | Code to evaluate when the server starts |
  236. #### `InstallMCPServer`
  237. | Option | Default | Description |
  238. |--------|---------|-------------|
  239. | `"EnableLLMKit"` | `Automatic` | Enable or disable LLMKit features in the context tools; `False` sets `LLMKIT_ENABLED=false` so they behave as unsubscribed with no subscription warnings |
  240. | `"EnableMCPApps"` | `True` | Enable or disable [MCP Apps](docs/mcp-apps.md) UI resources |
  241. | `"DevelopmentMode"` | `False` | Use local source files instead of installed paclet |
  242. | `"MCPServerName"` | `Automatic` | Override the config file key for the server entry |
  243. | `"ToolOptions"` | `<\|\|>` | Customize built-in tool behavior (see [docs/tools.md](docs/tools.md#tool-options)) |
  244. | `"VerifyLLMKit"` | `True` | Check LLMKit subscription requirements |
  245. ## Development
  246. See the [developer documentation](docs/README.md) for information on:
  247. - [Getting started](docs/getting-started.md) with development
  248. - [Writing and running tests](docs/testing.md)
  249. - [Building the paclet](docs/building.md)
  250. - [Adding new tools](docs/tools.md)
  251. - [Agent skills](docs/agent-skills.md) for distributing tools to AI coding agents
  252. - [Paclet extensions](docs/paclet-extensions.md) for third-party tool/prompt/server contributions
  253. - [Error handling](docs/error-handling.md)
  254. For AI agents working on this codebase, see [AGENTS.md](AGENTS.md).
  255. ## License
  256. This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
  257. ## Author
  258. Richard Hennigan (Wolfram Research)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.