Post tweets, reply, like, retweet, follow, DM & more - directly from your chat. Full X/Twitter automation for OpenClaw.
Hermes Agent users should use Hermes Tweet, the native Python plugin for the same Xquik API contract.
Search tweets, search tweet replies, post tweets, post tweet replies, export followers, look up users, upload media, download media, send direct messages, monitor tweets, deliver webhooks, and run giveaway draws through structured agent tools. Powered by Xquik, the all-in-one X automation platform.
openclaw plugins install @xquik/tweetclawOpenClaw resolves this package through ClawHub first and uses the exact npm package as a fallback when needed.
TweetClaw can be installed before credentials are configured. Until you add an API key or MPP signing key, the free explore catalog remains available and live API calls return setup guidance instead of failing plugin installation.
Agent-skill installers can also discover TweetClaw through Skills.sh. This installs the packaged SKILL.md for agents that use skills directories; use the OpenClaw command above for the plugin runtime.
npx skills add xquik-dev/tweetclawNote:
@xquik/tweetclawis the only official npm package. Any other scope (for example@intentsolutionsio/tweetclaw) is an unofficial redistribution and may ship stale metadata or outdated endpoint counts.
TweetClaw uses Xquik billing for account-backed automation, credit top-ups, and optional MPP pay-per-use reads. See Billing & Pricing for the current plans, eligible endpoints, and operation costs.
Get an API key at dashboard.xquik.com. Store it in an environment variable and configure TweetClaw to use it:
openclaw config set plugins.entries.tweetclaw.config.apiKey "$XQUIK_API_KEY"Security: Keep the key out of chats, docs, and shell history. Prefer the environment-variable command above so OpenClaw writes the secret to its local config without exposing it in the prompt.
Top up credits from the Xquik dashboard. TweetClaw does not create checkout sessions or charge saved payment methods from the agent.
MPP (Machine Payments Protocol) lets agents pay per API call without an account, API key, or subscription. 31 read-only endpoints. Create an MPP account with mppx account create. The signing key stays local and is only used to sign payment proofs.
npm i mppx viem
openclaw config set plugins.entries.tweetclaw.config.tempoSigningKey "$MPP_SIGNING_KEY"Security: Keep the signing key out of chats, docs, and shell history. Prefer the environment-variable command above so OpenClaw writes the secret to its local config without exposing it in the prompt.
MPP-eligible read endpoints include tweet lookup, tweet search, user lookup, user tweets, follower checks, articles, trends, quotes, replies, retweeters, favoriters, threads, user likes, user media timeline reads, communities, lists, users batch lookup, people search, followers, following, mentions, and verified followers. See the billing guide for current endpoint eligibility and costs.
Media downloads and gallery creation are not MPP-eligible. The user media endpoint returns a user's media-tweet timeline; it does not download files or create gallery links.
OpenClaw loads explore as the safe catalog tool. The live endpoint invoker, tweetclaw, is registered as an optional tool because it can perform paid reads, private reads, and write actions.
OpenClaw's local onboarding default is often tools.profile: "coding", which excludes external plugin tools from agent runs until they are explicitly allowed. If the agent can see the TweetClaw skill but cannot call the tools, add the tool names to tools.alsoAllow so you keep the normal coding tools and opt into TweetClaw.
openclaw config set tools.alsoAllow '["explore", "tweetclaw"]'Verify runtime registration after install or update:
openclaw plugins inspect tweetclaw --runtime
openclaw skills info tweetclawopenclaw config set plugins.entries.tweetclaw.config.baseUrl "https://xquik.com"
openclaw config set plugins.entries.tweetclaw.config.pollingEnabled true
openclaw config set plugins.entries.tweetclaw.config.pollingInterval 60Only change baseUrl for a self-hosted Xquik-compatible API. TweetClaw requires an HTTPS base URL with no embedded credentials.
TweetClaw uses 2 structured tools for the agent-safe endpoint catalog:
Search the API spec to find endpoints. No API calls are made.
You: "What endpoints are available for tweet composition?"
AI uses explore → filters spec by category "composition"
→ Returns matching endpoints with parameters and response shapes
Invoke catalog-listed API endpoints with structured path, method, query, and body fields. Auth is injected automatically - the LLM never sees your API key.
This tool is optional in OpenClaw. If your agent can see the skill but cannot call TweetClaw tools, add explore and tweetclaw to tools.alsoAllow so your normal tool profile stays intact.
OpenClaw approval prompts are enforced before write-like tweetclaw tool calls. Review the structured request before approving any post, delete, follow, DM, monitor, extraction, webhook, or profile-change action.
You: "Post a tweet saying 'Hello from TweetClaw!'"
AI uses tweetclaw → finds connected account, posts tweet
→ Returns { tweetId, success: true }
You: "Search tweets about AI agents"
AI uses explore → finds /api/v1/x/tweets/search
AI uses tweetclaw → calls the endpoint with auth
→ Returns tweet results
Instant responses, no LLM needed:
| Command | Description |
|---|---|
/xstatus |
Account info, subscription status, usage, credit balance |
/xtrends |
Trending topics from curated sources |
/xtrends tech |
Trending topics filtered by category |
When polling is enabled (default), TweetClaw checks for new events every 60 seconds and delivers them to your chat:
- Monitor alerts: New tweets, replies, quotes, retweets from monitored accounts
Set up a monitor first:
You: "Monitor @elonmusk for new tweets, replies, and retweets"
99 agent-callable endpoints across 9 categories. Dashboard-only account-admin, billing, support, and raw credential flows are excluded from the tool catalog and blocked at runtime.
| Category | Examples | Access |
|---|---|---|
| Account | Account status | Account-backed |
| Composition | Compose, drafts, writing styles, radar | Account-backed |
| Credits | Check balance | Account-backed |
| Extraction | 23 extraction tools, giveaway draws, exports | Account-backed |
| Media | Upload media via URL, authenticated media download, gallery links | Account-backed |
| Monitoring | Create monitors, view events, manage webhooks | Account-backed |
| Search, lookups, timelines, articles, trends, bookmarks, notifications | Account-backed or MPP where eligible | |
| X Accounts | List connected account handles for explicit user-selected actions | Account-backed |
| X Write | Post, reply, like, retweet, follow, remove follower, DM, profile, communities | Account-backed with approval |
Media download requires authenticated access and is not MPP-eligible. MPP only includes media-tweet timeline reads, not file download or gallery creation.
- Xquik Platform
- API Documentation
- Billing & Pricing
- Context7 Agent Docs
- OpenClaw Setup Guide
- Agent Workflow Guide
- Framework guides: Mastra, CrewAI, LangChain, Pydantic AI, Google ADK, Microsoft Agent Framework, n8n, Zapier, Make, Pipedream, Composio migration
- npm Package
- Hermes Tweet for Hermes Agent
- OpenClaw
MIT