X MCP guide
An agent skill by cursor, from cursor/plugins. Tags: api, documentation, security, setup, troubleshooting, validation.
What it does
ALWAYS read this when a user connects the X plugin or any X MCP, before using any X connection, and again on any X error. Do not call an X tool until this file has been read in the current turn. On first connect, confirm X tools are available, fetch get_usage_credits BEFORE any user-facing text, then send the congrats + capabilities message. Never tell the user to buy credits until that check returns ~$0 or a job would exceed the balance. If X is connected but tools are missing (tools=0, user-X-* not found), that is a setup failure — not a paywall. For encrypted X Chat / XChat DMs, also read skills/x-chat/SKILL.md. If chat tools or dm.read/dm.write are missing while other X tools work, tell them to reconnect the X plugin. Estimate the cost of every X call before making it and confirm with the user before anything expensive.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add cursor/plugins --skill 'X MCP guide'Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):
git clone --depth 1 https://github.com/cursor/plugins
cp -r plugins/third_party/x/skills/x-api-mcp-guide ~/.claude/skills/'X MCP guide'Safety box score
Grade F, high risk: 72 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.
This skill is a full X integration guide that can read timelines, DMs, calls, bookmarks, trends, and livestream controls, and it also bootstraps a separate helper repo for Chat decryption. The main risks are broad outbound actions on the user’s X account, handling/storing Chat PIN and Juicebox key material, and executing externally fetched code plus live API calls.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 3 (high) | `SKILL.md` tells the agent to run shell and Python commands such as `git clone https://github.com/xdevplatform/xchat-grokbot-helper.git`, `python3 -m venv ...`, `pip install -U pip chatxdk`, and `$HELPER $SCRIPT ... unlock-check`. |
| Network access | 3 (high) | The skill explicitly pulls code from GitHub and calls remote X/console endpoints, including `https://console.x.com/api/credits/pricing` and many `GET /2/...` MCP calls that exchange the user’s X data with remote services. |
| Filesystem reach | 2 (moderate) | It writes outside the project tree to `$HOME/xchat-lite`, persists `juicebox_config` as a mode-600 file, and manages saved key material on disk with `forget`/`--remember-key`. |
| Secrets & credentials | 3 (high) | The guide handles `CHAT_PIN`, says `Never echo it`, uses `juicebox` material for decryption, and can save the X Chat private key on this computer for later sessions. |
| Privilege & persistence | 0 (none) | There is no sudo, global install, service registration, shell startup editing, or OS-wide privilege change in the shipped files. |
| Instruction hijack surface | 2 (moderate) | The guide repeatedly overrides assistant behavior with directives like `Do not ask for confirmation`, `Do not search`, and `Do not look for a second skill`, which constrains normal review and confirmation behavior. |
| Irreversible actions | 3 (high) | It can send/reply to DMs, accept message requests, schedule/cancel calls, create/delete broadcasts and sources, and mute/unmute users on the owner’s behalf. |
| Transparency | 2 (moderate) | The skill depends on an external, unpinned helper repo (`git clone https://github.com/xdevplatform/xchat-grokbot-helper.git`) and a live pricing endpoint that are not included in the reviewed files. |
Before installing
- Read `SKILL.md` sections that clone and run `xchat-grokbot-helper`; confirm you trust code fetched from GitHub.
- Decide whether you want the skill to store X Chat key material under `$HOME/xchat-lite` on this machine.
- Verify you’re comfortable granting write access for X DMs, calls, and livestream actions, not just read access.
- If possible, test it first on a disposable or low-risk X account.
Rated 2026-10-20 by an automated review of 2 files from the skill's repository.
Source
- Repository
- cursor/plugins (all skills from this repository)
- Path
- third_party/x/skills/x-api-mcp-guide/SKILL.md
- Branch
- main
- Collection
- cursor
- Updated
- 2026-09-23
Related skills
- principle-boundary-discipline — Apply when wiring validation, error handling, or framework adapters.
- expo-data-fetching — Framework (OSS). Use when implementing or debugging ANY network request, API call, or data fetching.
- seo-technical — Technical SEO audit across 9 categories: crawlability, indexability, security, URL structure, mobile, Core Web Vitals, structured data, JavaScript rendering.
- sharp-edges — Identifies error-prone APIs, dangerous configurations, and footgun designs that enable security mistakes.
- agent-web-search-setup — Sets up working web search on an agent whose model backend cannot run it.
- amazon-ses — Guides Amazon SES onboarding for domain-based email sending.