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

  • Godot MCP NativeMCP server
    • The execute-script and execute-editor-script tools run GDScript in the editor and in the running game, and script modification writes project files
    • The HTTP server listens on port 9080 and by default needs no token, so protection has to be switched on by hand
  • Godot MCP ai-game.devMCP server
    • The agent writes and deletes project files, scenes and scripts, and script-execute compiles and runs arbitrary C# in the editor
    • By default the addon connects to the ai-game.dev cloud, and editor data passes through that service
    • Sign-in uses OAuth and the credential is saved on the machine for the editor to reuse
  • GolemBotCLI
    • Chat members get an agent that runs commands and reads files in its working folder
    • Messenger tokens and model provider keys sit in the bot configuration
  • Google Calendar MCP Server: calendar for ClaudeMCP server
    • The agent reads your schedule, creates, changes and deletes events and adds attendees
    • The OAuth2 client secret must live in the client config, and sign-in opens access to the Google account's calendars
  • Google Docs MCP: Docs, Sheets, Drive, Gmail and CalendarMCP server
    • The agent gets rights to read and write documents and Drive, send mail (gmail.modify) and change calendar events (calendar.events)
    • The tools delete data: deleteFile can remove a file permanently, and deleteDraft and deleteEvent are irreversible
  • Google MCP SecurityMCP server
    • The servers read investigation data, security events and cases and pass them into a language model's context
    • The SOAR server runs orchestration integrations that act in external systems such as OKTA
    • Google credentials and the VT_APIKEY and SOAR_APP_KEY keys open access to an organization's security products
  • Google Play Developer MCPMCP server
    • Can publish releases to production and change the app listing
    • Answers reviews and manages subscriptions on behalf of the developer account
    • The server can delete app metadata, and the README advises minimal roles
  • Google Workspace Extension for Gemini CLIPlugin
    • Reads, changes and deletes the account's mail, documents and events
    • Exposed to prompt injection from incoming mail and documents
  • Google Workspace MCP ServerMCP server
    • Sends email and edits documents and calendars on the user's behalf
    • Stores OAuth tokens with broad account access
  • Goose SkillsSkill
    • Runs local scripts and sends requests through a third-party paid API
    • Uses an account token and spends credits
    • Some workflows prepare and run outreach
  • Grafana MCP ServerMCP server
    • Can change dashboards, alerts and incidents in production Grafana
    • Holds a token with access to metrics and logs
  • GraphJin MCPMCP server
    • Lets the agent run queries against a working database, including writes where that is permitted
  • Graphicle: code graph and project memory for agentsMCP server
    • Setup edits agent configuration, adds Git hooks and turns on session capture in the project
    • The index covers code, documents and conversations with the agent, and with answers turned on, excerpts go to the chosen model provider
  • GreenfieldPlugin
    • Agents read all analyzed code and can take binaries apart or run the target program in a container
    • Writes many files into the workspace and spends many tokens
  • Grok BuildCLI
    • Runs shell commands and edits project files
    • Installs plugins and connects MCP servers, including ones found in Claude Code settings
    • Sends code and prompts to xAI servers
  • Growth LabSkill
    • Service keys and WeChat credentials sit in local settings, and the agent works with them from the working folder
    • Can create pages, submit URLs to a search engine and sync drafts to WeChat
    • Collectors for third-party platforms drive a browser and can lead to limits or an account ban
  • GuaardvarkPlugin
    • Screen agents click and type in a real browser on a separate Ubuntu desktop
    • Once desktop, shell and browser tools are allowlisted, any agent connected over MCP gets them
    • The studio stores and processes voice, video and files on your machine
  • HACK.SKILLSSkill
    • Offensive methods can cause harm without the system owner's permission
    • Requires strictly authorized use and legal responsibility
  • HAPICLI
    • Grants remote access to the work machine and agent sessions from a phone
    • By default traffic passes through a shared third-party relay server
  • HEY CLICLI
    • The agent reads the mail, contacts and calendar of your HEY account
    • The agent saves drafts in your name, and it can run `hey draft send`, which sends a draft, through the terminal
  • HH MCP ServerMCP server
    • Can submit job applications on your behalf with a cover letter
    • Uses browser automation on a real hh.ru account, which may go against the service's rules
  • HH.ru MCP Server: browser auto-applyMCP server
    • Can automatically submit applications with an AI-generated letter without a mandatory manual review of each one
    • Uses Playwright stealth settings explicitly aimed at evading hh.ru's anti-bot defenses, which can lead to an account ban
  • HOL GuardCLI
    • Writes its own hooks into Claude Code, Codex, Cursor and other agents' settings and puts package-manager shims on PATH
    • A background process sees the full text of every command, file read and MCP call before it runs and keeps a log in ~/.hol-guard
    • After hol-guard connect, decisions sync to Guard Cloud; remote commands to the device are a separate opt-in
  • HTTP 1C MCP gateway with a dashboardMCP server
    • Claims BSL code writing and access to the event log of connected infobases
    • Gives access to 1C data and metadata, worth starting with a test infobase
  • Habr Career MCP serverMCP server
    • Uses a session cookie that grants full access to the Habr Career account
    • Writes to the profile through site forms rather than the official API
  • Habr MCP server for reading and publishingMCP server
    • Stores the user's browser session cookie on the server, even if encrypted
    • Can create, edit and delete article drafts on the author's behalf
  • HalofyPlugin
    • The stop hook sends the session transcript to the server, and it may contain code, data and secrets
    • The agent key in an environment variable opens read and write access to organizational memory
    • The plugin installs an MCP server and hooks that fire at the start and stop of every session
  • Hardened VK Ads MCPMCP server
    • With VK_ADS_ENABLE_WRITES on, it can create, change and delete campaigns and spend ad budget
    • File uploads to campaigns reach out to external addresses, even though SSRF protection is in place
  • Harmonist: agent pack with enforced hooksPlugin
    • Hooks run Python and bash scripts on every subagent dispatch, file edit and session stop
    • The integration script writes AGENTS.md, memory and hooks into the project, and memory holds notes about your decisions
  • Harness AnythingCLI
    • Drives installed programs through COM and creates, edits and overwrites documents, presentations and graphic files
    • Before building presentations the instructions advise force-closing the `wps.exe` process, which discards unsaved work in it
    • The `cli-anything-wps` tool installs straight from a GitHub repository and its code runs with your Windows privileges
  • Harness MCPMCP server
    • Starts other coding agents that run commands and write files in the given folders and in new projects
    • May send work to your other machines, where a session acts with that machine's user rights
    • Installing Harness means running a script downloaded over the network, and on Linux it installs packages with administrator rights
  • HarnessKitCLI
    • The app reads and changes the configs, hooks, MCP and memory of every agent it finds
    • Deploying an extension to other agents adds hooks and servers that will run when agents work
    • Installation runs a `curl | sh` script, and web mode opens a port on the machine
  • Hass-MCPMCP server
    • Controls real home devices, including lights, appliances and locks
    • Can restart Home Assistant and call any of its services
    • Edits dashboards and handles a long-lived access token
  • HeadHunter MCP ServerMCP server
    • Sends job applications on the user's behalf
    • Reads resumes containing personal data
    • Holds app secrets and OAuth tokens
  • HeadroomMCP server
    • All agent traffic to the model provider, including API keys, passes through the local proxy
    • wrap changes agent configuration and registers extra MCP servers
    • learn writes changes into CLAUDE.md and AGENTS.md
  • Hermes AgentCLI
    • Installs via curl piped to bash and downloads binaries
    • The agent runs shell commands and acts autonomously on schedules
    • Stores provider keys and messaging tokens
  • Heroku Platform MCP ServerMCP server
    • Runs SQL and kills processes in Postgres databases
    • Scales, transfers and deploys apps
    • Runs arbitrary code on one-off dynos
  • Hex-Rays IDA MCPPlugin
    • The execute_python tool runs arbitrary Python inside the IDA database and save_database writes changes to disk
    • The agent opens any binary by path, including potentially malicious samples
  • HexStrike AI MCP AgentsMCP server
    • Runs offensive tools against real targets
    • Actions may be illegal without the target owner's authorization
    • Executes system commands with your privileges
  • Higgsfield Claude Skills: 19 skills for video and UGC adsSkill
    • The skills drive a Playwright browser under your Higgsfield account: they type text, pick files and press buttons, including the one that starts generation
    • JavaScript snippets run on the service's pages, and you sign in to your account in the same Playwright window
    • The `ugc-hot-girl` skill writes prompts for photorealistic fictional characters in ad reviews: such a clip must be labeled as advertising and as AI-generated, and must not be passed off as a real customer review
  • HindsightMCP server
    • Requires LLM provider keys and handles secrets
    • Sends memory content to an external LLM provider for processing
    • Runs a persistent server and database, and the CLI installer reads git history and past sessions
  • HiveCLI
    • Workers run in a mode without approval prompts and execute shell commands with your account's rights
    • There is no sandbox or multi-user authentication, and every agent gets access to the whole working directory
    • The optional remote access from a phone exposes the full Hive interface with the same rights as the local browser
  • HivemindPlugin
    • Sends prompts, tool calls and responses from every session to an external cloud
    • Installs hooks into every detected assistant
  • Home Assistant Vibecode AgentMCP server
    • The agent controls the home and automations
    • A long-lived access token is needed
    • A third-party project with no releases
  • HomeButlerCLI
    • The write and destructive tools manage containers, backups and Proxmox guest power
    • Access to servers goes through SSH keys, and logs and container configs may contain credentials
    • The install commands deploy third-party apps on the server
  • HomunculusPlugin
    • Hooks run the `observe.sh` and `on_stop.sh` scripts on every prompt, tool call and session stop
    • A background agent reads those records on its own, and the plugin creates new commands, skills and agents in the project
  • HugAgentOSCLI
    • Runs code in a sandbox and carries out tasks the agent decides on
    • Hosts a network service with a database and file storage
    • Connects external models and MCP tools and accesses the network
  • Hugging Face SkillsSkill
    • Uploads and deletes data in Hub repos on the user's behalf
    • Launches paid jobs on cloud GPUs and AWS
  • Huntflow MCP serverMCP server
    • The server reads candidate resumes and profiles with a Huntflow token, which is personal data
    • Refreshes the access token on its own with the refresh token
    • There are no tools that write to Huntflow
  • Hyperbrowser MCPMCP server
    • Browser agents click buttons and fill in forms on sites themselves, and persistent profiles keep logged-in sessions
    • All pages, requests and data pass through the Hyperbrowser cloud under your key
    • In SSE mode without the `--sse --serve` flags the key check is off

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.