SkillFoxx

Type a name, a task or a service

Match with Foxx AI

Access levels of skills and MCP servers

The access level of every catalog entry and the reasons behind it. The level shows what a tool adds to what the agent can already do. It does not rate code quality.

Found: 1417

  • SSH MCP Server (classfang): remote commands over SSHMCP server
    • The agent runs commands on a remote server with the chosen user's rights and can change or delete files
    • The login password or key sits in the client configuration, and the examples pass the password as a plain argument
  • STS2MCPMCP server
    • The mod opens an API on localhost through which an outside program reads and controls the game
    • The agent can switch and delete profile slots, with their progress, using delete_profile
    • The mod loads into the game client, so its failures can affect saves and startup
  • SWE-AFWorkflow
    • It launches hundreds of agent calls that edit code, commit and merge branches in your repository
    • With `GH_TOKEN` it clones private repositories, pushes branches and opens pull requests in your name
    • It uses provider keys and a Claude or Codex subscription, so a long build spends API balance or plan limits
  • Safe DOCXMCP server
    • Edits and saves DOCX files on your disk
    • Converts and overwrites documents
    • A young project, version 0.23.0
  • Salesforce DX MCP ServerMCP server
    • Changes metadata and data in Salesforce orgs
    • Uses local Salesforce CLI authorizations
  • Salesforce Skills LibrarySkill
    • Deploys metadata and runs code in Salesforce orgs, including production
    • Works with access permissions and platform security settings
  • Sanity Agent ToolkitPlugin
    • MCP can modify documents and datasets in Sanity projects
  • Saymon RU Data API: paid Russian data endpointsMCP server
    • The agent can automatically spend real USDC cryptocurrency on every call without separate confirmation
    • The wallet connected for payment is directly accessible to the tool
  • Sber API MCP for corporate clientsMCP server
    • The server works with ruble payment orders on the company's corporate account
    • Access exposes account statement data, which requires strict permission control and a bank contract
  • Sberbank business API MCP serverMCP server
    • The create_payment tool creates a real payment order and moves money from the company account
    • The implementation is unverified against a live bank environment; test on an account with a minimal limit
  • ScrapeAtlas CLICLI
    • The customer key sits in a config file as plain text, and an agent holding it spends paid credits and can drain the balance with a burst of requests
    • Requests and responses pass through the vendor's server, and answers contain data about accounts and people that you need the right to collect, within each platform's terms
  • ScrumboyMCP server
    • A project token lets the agent create, change, archive and delete projects and tasks
    • The admin_* tools work with an instance's users and roles
    • The server takes tokens and session cookies over HTTP, which are easy to intercept without TLS in front
  • Search Console MCP: Google, Bing and GA4 data for agentsMCP server
    • The agent gets access to Google, Bing, GA4 and AdSense statistics, including revenue and account data
    • With the separate write scope, tools submit and delete sitemaps, add and remove properties and submit URLs for indexing
  • Self Media SkillsPlugin
    • The skills create WeChat drafts and, once authorized, upload video to accounts on YouTube, Douyin, WeChat Channels and Xiaohongshu
    • WeChat needs the application secret WECHAT_APP_SECRET in the environment
  • SendPulse MCP serverMCP server
    • Can launch real email and SMS campaigns on behalf of the account
    • Updates user data in the SendPulse database
  • SendPulse MCP serverMCP server
    • Can send an email campaign to the entire mailing list in one call with no confirmation
    • The OAuth2 credentials grant access to the account's campaigns and statistics
  • SendPulse MCP server for WhatsApp and SMSMCP server
    • Can send messages and SMS to real contacts and launch flows without confirmation
    • Can delete contacts and cancel scheduled campaigns
  • Sentry MCPMCP server
    • The token grants write access to Sentry projects, teams and events
    • Error data may be sent to a third-party LLM provider for search
  • Sentry for AISkill
    • The plugin connects the Sentry MCP server, so the agent reads errors and events of your environment, which can contain user data
    • The skills create alerts and resolve issues in Sentry, and sentry-debug-issue edits project code for the error it finds
    • The sentry-instrument skill adds the SDK and monitoring configuration to application code
  • SerenaMCP server
    • Edits project files
    • Includes a shell command execution tool
  • ServiceNow MCP Server (NowAIKit)MCP server
    • ServiceNow credentials sit in the server environment and give access to incidents, users and configuration
    • Enabled write flags let the agent create and change records, CIs and server-side scripts on the instance
    • Incident data goes to the client's model, and on a production instance a scripting mistake affects the whole company
  • SessionBridge: a temporary bridge to 1C over RDPMCP server
    • Gives the agent read and write access to the client's 1C infobase through a temporary outbound tunnel
    • The session token is passed manually and stays valid until the session is stopped
  • Shared Sessions for Claude CodePlugin
    • A guest can edit files, run commands and use the network through your machine if you approve the request
    • The session transcript and shared files pass through the room server without end-to-end encryption
    • Moving to the cloud sends your Claude token and project files to a server container
  • ShellGPTCLI
    • The --shell mode and function calling let the model suggest and, once you confirm, run any command as your user
    • Prompt text and piped data go to the chosen model provider, and the API key is stored in the ~/.config/shell_gpt/.sgptrc file
  • ShepCLI
    • Runs the agent with permission-bypass flags by default, so it executes commands and edits files without asking
    • Pushes branches and opens pull requests on your behalf through GitHub CLI
    • Sends usage metrics to the project team until you turn them off
  • Ship SafeCLI
    • Hooks are written to ~/.claude/settings.json and see every file write and Bash command in Claude Code
    • The fix agent changes project files, and with `--branch --pr` it creates a branch and opens a pull request
    • The AI modes pass code to a provider, and `--verify` sends found keys to their providers
  • ShippieCLI
    • Requires a model provider key as a secret and writes comments directly into the pull request
  • Shodh-MemoryMCP server
    • Stores the content of your sessions and decisions on disk and starts a local HTTP server on port 3030
    • Claude Code hooks record session context on their own without being asked
    • First run downloads a binary and models from the network
  • Shopify AI ToolkitPlugin
    • Runs store operations through Shopify CLI
    • Sends queries, code and user messages to Shopify by default
  • Shturman: Direct MCP with a hard red lineMCP server
    • direct_apply actually changes a live account with real money
    • Creates Audience geo-segments and Metrica goals that affect future campaigns
  • SimUtil: MCP for emulators and simulatorsMCP server
    • The agent taps, types and captures the screen of devices, including real phones signed in to your accounts
    • Recordings and screenshots can capture personal data open on the device
  • SketchUp MCPMCP server
    • The eval_ruby tool is on by default and runs arbitrary code inside SketchUp
    • Creates and changes the geometry of your model
    • Works with a live program on your computer
  • SkillClawCLI
    • Proxies the agent's requests and rewrites its skills
    • With a shared server session data goes to shared storage
    • A research project with no promised results
  • SkillHub (iFlytek)CLI
    • The server install script is downloaded from an external address
    • The service serves skills that agents execute
  • SkillsVoteSkill
    • The hosted variant sends the rewritten task and a per-subtask action summary to api.skills.vote, and keeps the API key in a .env file inside the skill folder
    • Downloads third-party skills from GitHub into the project, and the agent then follows their instructions and runs their scripts
    • Runs scripts through `uv run` and may write a GitHub token into the same .env file
  • Skillz: MCP server for Claude-style skillsMCP server
    • The server can run helper scripts from skills on your machine
    • Skills from third-party sources may contain malicious code, and the author advises running the server in a sandbox or container
  • Slack MCP ServerMCP server
    • Reads private and work conversations
    • Uses Slack tokens and session cookies
    • Can post as the user when enabled
  • Slack MCP and Skills Plugin: the official Slack pluginPlugin
    • The agent reads channels and private messages and sends and schedules messages in your name
    • Permissions follow your Slack account, and OAuth sign-in opens access to workspace data
  • SmartChainPlugin
    • The assistant can call real smart-home devices through the Assist API: locks, lights, climate
    • It can analyze home camera snapshots on request
    • It sends home data to whichever cloud LLM provider is configured
  • Sno StationCLI
    • Hooks go into every agent found on the computer unless you name one with the `--harness` flag
    • Memory stores the content of your agent sessions, and in REM Enhanced mode the memory text goes to Sno's cloud
    • The nightly loop proposes edits to agent skills, and they take effect once you accept them
  • Sno Station SkillsSkill
    • The installer downloads over curl and runs in the shell at once, then puts the sno program and skills into every agent it finds
    • A change of shift wakes a second agent and hands it work on your code from a given commit; it needs background processes, tmux and logins to both CLIs
    • The nightly loop reads the day's Claude Code and Codex sessions, which hold your code and conversation
  • Solana MCP by VybeMCP server
    • Works with crypto data and assembles unsigned transactions for swaps, account closes and withdrawals
    • Calls an external paid API and requires authentication
  • SolidCraftMCP server
    • Can read, overwrite and delete files in allowed folders: the open design, the recovery folder and paths from `--allow-path`; the `--allow-any-path` flag lifts the limit
    • Runs the engine on the GPU and executes any CAD command through `execute` and `batch`, while `new_design` and `open` discard unsaved work
  • SolidPilotMCP server
    • The server drives a live SolidWorks session through COM: it opens, changes, saves and closes documents
    • edit_feature deletes and suppresses feature tree items, and save_analysis writes files to a .solidpilot folder next to your parts
  • SolidWorks MCP in PythonMCP server
    • Runs operations in your SolidWorks and changes the model
    • Works with a live licensed program on your computer
    • Installed from sources with no ready package
  • SolidWorks MCP in TypeScriptMCP server
    • Runs COM calls and VBA macros in your SolidWorks
    • The author has not verified most tools on a live program
    • Creates, changes and exports model files
  • SoupCLI
    • It runs training, installs packages and downloads weights from Hugging Face; a GGUF export may call pip unless `--no-install-deps` is given
    • `soup push` and `--push-as` send models and checkpoints to your Hugging Face account with a token
    • Running `--cloud modal` or `--cloud lambda` only prepares a plan, while `--cloud-submit` sends a paid job to the cloud
  • SourceCraft MCP and CLI: repos and CI/CDMCP server
    • It can create and delete repositories, merge pull requests and trigger CI/CD workflows
    • The personal token grants access to repositories and organizations under the token owner's rights
  • Spec-Driven DevelopPlugin
    • Creates, groups and closes Issues, milestones and labels in your GitHub repository through the gh CLI
    • Parallel lanes (Tier 2) write code in isolated worktrees and commit fixes directly to lane branches themselves
    • The execution phase merges several tasks into one PR and pushes changes to a shared integration branch
  • SpellbookPlugin
    • `install.sh` clones the repository and creates links in ~/.claude and ~/.agents, and the one-line install runs a script from the network
    • The library has skills that change server settings (SSH, firewall) and clean disks, and `push-all` commits and pushes all changes
    • Some skills use external CLIs, accounts and credentials, for example for publishing content

How we set the access level

Methodology

The editors set the level and the reasons from the README, the way the tool runs and the permissions it needs. Reading project files the agent can already see does not raise the level. There is no automatic code audit yet.