x-use
x-use connects an MCP client to X through an authenticated browser session for research, inbox workflows, reviewed publishing, outreach, and account operations.
Research profiles, posts, threads, home feed, bookmarks, notifications, inboxes, conversations, and analytics.
Search tweets and profiles, read tweet/thread context and media, and prepare replies.
Draft, review, approve, reject, queue, and publish posts, replies, and multi-post threads.
Schedule and process queued posts/engagement, with draft approval and recovery flows.
Send reviewed DMs, follow/unfollow, bookmark/unbookmark, like/unlike, retweet/undo, edit/delete own tweets, and manage supported inbox controls.
Maintain local leads and campaigns with personalization, opt-out suppression, and campaign preview/approval.
Manage multiple accounts, cookies, proxies, sessions, account health, safety pauses, and runtime diagnostics.
x-use — MCP server for X (Twitter)

Connect your AI client to X research, inbox reads, and reviewed publishing.
x-use is an open-source Python Model Context Protocol (MCP) server and CLI for X (Twitter). Connect Claude, Codex, Cursor or another local MCP client to profile research, inbox management and reviewed publishing. It uses your authenticated browser session, supports multiple accounts and runs locally without official X API credentials.
Quickstart · Documentation · Tool reference · Discussions · Contribute
What can you do with x-use?
Research before replying: search X, inspect a profile, read posts and bounded thread context, and prepare a relevant reply.
Work through your inbox: inspect visible conversations, unread messages, requests and notifications; prepare individual messages and supported conversation actions for review.
Publish with context: stage posts or multi-post threads, inspect the exact text, and approve the draft when it is ready.
Organize outreach locally: maintain leads, opt-outs and campaign drafts with per-account limits and delivery tracking.
Your AI client supplies the reasoning and text. An optional OpenAI-compatible LLM configuration supports server-side generation; ordinary explicit-text MCP workflows do not need that extra key.
Related MCP server: Spectre
Quickstart
Requires Python 3.11.5+ and Chromium. The CI matrix covers Python 3.11–3.14 on Windows, macOS and Linux.
Version 3.3.0 provides 82 tools for research, inbox management, reviewed publishing and account operations. Upgrade the package and restart your MCP client to load the current tools.
1. Install the server and browser
python -m pip install --upgrade x-use-mcp
python -m patchright install chromium2. Choose a private data directory and initialize
Windows PowerShell:
$env:X_USE_HOME = "$env:LOCALAPPDATA\x-use"
x-use init
x-use doctormacOS / Linux:
export X_USE_HOME="$HOME/.local/share/x-use"
x-use init
x-use doctorThe wizard configures your account and imports your own X session cookies from a local JSON export. Keep cookies and inbox PINs private. Use the same absolute data directory in your client configuration below; shell environment settings alone do not configure a desktop client.
3. Connect your MCP client
Use this stdio configuration in clients that accept an mcpServers object, such as Claude Desktop:
{
"mcpServers": {
"x-use": {
"command": "/absolute/path/to/x-use",
"args": ["mcp"],
"env": {
"X_USE_HOME": "/absolute/path/to/x-use-data"
}
}
}
}Replace both paths. On Windows, use the installed x-use.exe path and forward slashes in JSON, for example C:/Users/you/AppData/Local/x-use for the data directory. Locate the executable with Get-Command x-use in PowerShell or command -v x-use on macOS/Linux.
Claude Code, Codex, Cursor and Windsurf have their own registration locations or commands. Follow the client setup instructions, then reload the client if required. Ask it to call list_accounts before starting browser work.
4. Try a first workflow
With draft mode enabled, ask your client:
Check my account health, research the profile I specify, and summarize its recent visible posts. Prepare one relevant reply draft and show me its ID and exact text. Wait for my approval before publishing.
The flow is get_account_health → get_profile_context → prepare_reply → reply_to_tweet → get_draft. The reply tool stages a draft under the default policy; prepare_reply supplies context. Use approve_draft only when authorized.
More workflow examples cover inbox triage, outreach and threads.
git clone https://github.com/ihuzaifashoukat/x-use.git
cd x-use
python scripts/setup_uv.py --devOn Windows you can use py -3 scripts/setup_uv.py --dev; on macOS/Linux use python3 if needed. Setup installs uv when missing, synchronizes locked dependencies into .venv, installs matching Chromium, creates missing sample configuration and runs diagnostics. Activate the environment or use its executable directly. Then initialize your account.
For assisted installation, use the setup prompt or install the root setup skill:
npx skills add ihuzaifashoukat/x-useThe Claude Code plugin includes the bundled skills and MCP configuration guidance.
Capabilities
Workflow | Representative tools | Details |
Profile and post research |
| Visible profile, post, media and bounded thread evidence |
Inbox and notifications |
| Inbox/request folders, read-state evidence and partial message history |
Publishing and inbox controls (3.1.0) |
| Reviewed changes; supported controls and limitations |
Reviewed publishing |
| Persistent drafts and resumable thread progress |
Individual outreach |
| Message and follow drafts require individual approval |
Leads and campaigns |
| Local records, suppression and bounded batches |
Account readiness (3.1.0) |
| Operational score, evidence and recovery advice, plus explicit proxy connectivity checks |
Scheduling and recovery |
| Durable budgets, explicit recovery and uncertain-outcome tracking |
Version 3.3.0 includes 82 tools, seven prompts, four resource patterns and seven bundled skills. 3.1.0 added runtime diagnostics, campaign previews, bookmarks, reviewed post edits/deletions, engagement reversal and inbox controls to the 70 tools in 3.0.0; 3.3.0 adds quote posts to post_tweet and keeps the same tool set. See publishing and inbox controls for supported layouts and testing limits.
The MCP guide documents every tool, prompt and resource. Validation evidence separates local protocol checks, synthetic browser fixtures and authorized live-account checks.
For a uvx client, use uvx x-use-mcp or pin the release with
uvx x-use-mcp==3.3.0. The x-use-mcp executable starts MCP without arguments;
the existing x-use mcp command remains supported. Clients pinned to 3.0.0
must use uvx --from x-use-mcp==3.0.0 x-use mcp.
How execution works
The current MCP runtime uses async Patchright Chromium with separate account contexts. Your configuration, cookies, drafts, queue and action journals live under X_USE_HOME. AI-client and optional model-provider data handling depends on the client and provider you choose.
The protocol implementation uses the MCP Python SDK v2 line (mcp>=2.3,<3). It negotiates handshake protocol versions 2024-11-05 through 2025-11-25 with existing clients.
Draft mode is enabled by default for ordinary publishing. Messages, follows and thread starts require draft approval. Direct publishing mode, explicit queue processing, opt-in auto-drain and the legacy batch runner have separate execution paths. Review the configuration reference before changing these settings.
The development checkout allows one MCP server process per X_USE_HOME, preventing concurrent servers from using stale draft and queue state. Close the existing connection before opening the same data directory in another client.
Limits to understand
Browser reads are bounded by visible X content. Followers, inboxes and threads are not complete exports; unavailable evidence remains unknown.
Opening conversations or notifications can change their native read state.
Encrypted inboxes may need the account owner's PIN again after a browser-context restart.
Analytics currently reports availability and the observed Premium gate; it does not provide a complete analytics dashboard.
X layout changes, expired sessions, account challenges and platform limits can interrupt work. Local controls do not guarantee protection from account restrictions.
The legacy Selenium batch engine (
x-use run/run_cycle) executes separately from MCP draft review. Community posting is not currently verified in the async runtime.
See browser behavior and recovery and responsible-use guidance.
Frequently asked questions
Does x-use require an X API key?
No. It operates visible X pages through a browser authenticated with your session cookies. It does not require official X API credentials. Optional server-side text generation may require an LLM provider key.
Which MCP clients can use x-use?
Clients that launch local stdio MCP servers can use x-use. Setup examples cover Claude Desktop, Claude Code, Codex, Cursor and Windsurf. Client-specific discovery and reload behavior varies.
Does x-use automatically send every drafted message?
No. Message drafts require individual approval. Other write behavior depends on draft mode, queue execution and the legacy runner; check the configured policy before requesting actions.
Can I manage multiple X accounts?
Yes. Accounts have separate cookies, browser contexts and policy settings, with optional proxy routes. Browser capacity is bounded and depends on your host and configuration.
Can I edit posts or manage my X inbox?
Version 3.1.0 added reviewed edits and deletion of your own posts, bookmarks, unfollowing, and supported conversation request, mute/unmute and deletion actions. X determines edit eligibility. Inbox history is partial, and per-message editing/deletion, attachment sending and group administration are not implemented. See the controls guide for exact arguments and limits.
Is x-use free and open source?
The project is MIT licensed with no project license fee. AI-client subscriptions, optional model providers, proxies and hosting may have their own costs.
Is x-use affiliated with X?
No. x-use is an independent community project. X and Twitter are names used to identify the platform it works with.
Was this project called twitter-automation-ai?
Yes. The project was renamed to x-use during the v2 relaunch. See the changelog for its history.
Documentation and community
Start with the documentation index. It links setup, workflows, tools, configuration, presets, architecture and troubleshooting. X-Use is maintained by Huzaifa Shoukat and welcomes community contributions.
Found a reproducible problem? Report a bug. Have a workflow to share? Start a discussion. Useful first contributions include clearer setup instructions, fictional browser fixtures, platform fixes and example presets.
See CONTRIBUTING, the Code of Conduct and security policy. If x-use helps your work, a GitHub star helps other developers discover it.
Star history
License
MIT.
Available Tools
82 toolsadd_accountA
Add an account to config/accounts.json (validated, backed up,
atomic). cookie_file is a path ON THIS MACHINE to an exported
x.com cookies JSON; it is validated and copied beside the selected
accounts configuration as <account_id>_cookies.json. Cookie values never pass through
the tool call. persona is freeform markdown describing the
account's voice/engagement style (max 4000 chars).
Inputs: account_id (required), proxy, persona, is_active, +2 more. Returns: result
add_campaign_leadsAIdempotent
Add 1-100 local leads from this account to a campaign. Repeated ids are ignored. The whole batch fails if any id belongs to another account. Adding a lead does not make an opted-out or closed lead eligible.
Inputs: account (required), lead_ids (required), campaign_id (required). Returns: result
add_proxyAIdempotent
Add a proxy URL to a named pool (created if new) in settings.json. Validated, deduped, backed up, atomic. Credentials are masked in the response. Assign it to accounts via update_account(proxy="pool:") or a direct URL.
Inputs: pool (required), proxy_url (required). Returns: result
approve_draftADestructive
Execute a pending draft created by a write tool. Runs the exact same execution path as direct mode (pacing, dedup, metrics). A draft can be approved exactly once; unknown or consumed ids return an error.
Inputs: draft_id (required). Returns: result
bookmark_tweetA
Prepare a bookmark draft for one exact post. Always requires approve_draft.
Inputs: account (required), tweet_url (required). Returns: object
cancel_queued_actionAIdempotent
Cancel a pending or failed queued action. Done, processing, and already-cancelled items cannot transition.
Inputs: queue_id (required). Returns: result
cancel_threadAIdempotent
Cancel remaining thread segments locally. Retains confirmed URLs and uncertain evidence; never deletes posts or clears safety uncertainty. An active send must finish before cancellation can be accepted.
Inputs: run_id (required). Returns: result
close_sessionBIdempotent
Close one account's browser context; retain the operator's cookie file.
Inputs: account (required). Returns: object
continue_threadA
Resume an approved run only at a known unsubmitted segment after spacing/budget/preflight recovery. Uncertain segments are never resent.
Inputs: run_id (required). Returns: result
create_campaignA
Create a local draft campaign. Personalize using {handle}, {name}, {display_name}, or {company}; include a recipient name/handle placeholder. {name} falls back to @handle. Notes cannot be templated. Nothing sends.
Inputs: name (required), account (required), message_template (required). Returns: result
delete_tweetA
Prepare permanent deletion of your own exact post. Supply its current text (empty for media-only). Always requires approve_draft; cannot be undone.
Inputs: account (required), tweet_url (required), expected_text (required). Returns: object
draft_post_variationsA
Generate count distinct takes on a topic with the server-side LLM
(persona-aware) and stage each as a post DRAFT. Nothing is posted,
approve one and reject the rest. Requires the 'llm' block in settings.
Inputs: topic (required), account (required), count. Returns: result
edit_tweetA
Prepare a text edit of your own post. Supply its current exact text and replacement; configure self_handles. X must expose Edit post; eligibility/time limits apply. Always requires approval. Media replacement is unsupported.
Inputs: text (required), account (required), tweet_url (required), expected_text (required). Returns: object
engageA
Engage with tweets found via keyword search. actions is a subset
of ["like", "retweet"] (default ["like"]); every candidate passes the
analyzer relevance gate; max_actions is hard-capped by the account's
per-run config caps. In draft mode (default) each planned action
becomes an individually approvable draft and nothing executes.
Inputs: account (required), keywords (required), actions, max_actions. Returns: result
follow_profileA
Prepare a follow draft for one X profile; approve_draft performs the action.
Inputs: account (required), profile (required). Returns: object
generate_and_postA
Generate a post about topic with the configured LLM and post it.
In draft mode (default) the generated text is stored in the draft's
payload/preview so a human reviews the actual content before anything
is posted.
Inputs: topic (required), account (required). Returns: result
get_accountARead-only
Show one account's masked config (no cookies, no password, proxy credentials masked) plus cookie-file status. Read-only, never starts a browser.
Inputs: account (required). Returns: result
get_account_analyticsBRead-only
Read the signed-in owner's X analytics availability. Requires X Premium when the observed paywall says so; unsupported layouts return no metrics. Distinct from local action metrics and public post counts.
Inputs: account (required), limit. Returns: object
get_account_healthARead-only
Read-only health snapshot for one account: config presence, cookie-file validity, metrics counters, warm-session state, queue depth, pending drafts, and a transparent local readiness score out of 100. Unknown evidence earns no points; this does not predict X enforcement or establish live authentication. Never starts a browser. A config that fails validation is REPORTED (config.valid=false with the error), not raised, surfacing that is the point of a health check.
Inputs: account (required). Returns: result
get_account_safetyCRead-only
Read local limits, pause reason, and writes needing outcome reconciliation.
Inputs: account (required). Returns: object
get_bookmarksARead-only
Read bounded posts in the default bookmarks feed. Partial results; bookmark folders are not traversed.
Inputs: account (required), limit. Returns: object
get_campaignCRead-only
Read this account's campaign and a membership page. limit 1-100; members_next_offset identifies the next page of local outcomes.
Inputs: account (required), campaign_id (required), limit, offset. Returns: result
get_campaign_summaryBRead-only
Read local membership/delivery counts and opt-out suppression count. These are recorded outcomes; this tool never checks X or sends work.
Inputs: account (required), campaign_id (required). Returns: result
get_conversationCRead-only
Read bounded visible conversation context and observed request/reply state. Cursor stays within the visible window; opening may clear X unread state. Does not accept requests.
Inputs: account (required), conversation_id (required), limit, before_message_id. Returns: object
get_conversation_controlsCRead-only
Inspect visible request and conversation-menu controls. Opening may mark read; unsupported layouts return an error, never guessed capabilities.
Inputs: account (required), conversation_id (required). Returns: object
get_draftARead-only
Show one draft by id. Read-only.
Inputs: draft_id (required). Returns: result
get_home_feedBRead-only
Read posts currently visible in the home feed; no engagement actions.
Inputs: account (required), limit. Returns: object
get_inboxBRead-only
Read visible inbox/requests/other summaries with all/unread/read filtering. Native Unread selection is verified when available; never opens or accepts a conversation.
Inputs: account (required), limit, folder, inbox_filter, +1 more. Returns: object
get_leadBRead-only
Read one local lead belonging to this account, including its notes.
Inputs: account (required), lead_id (required). Returns: result
get_metricsARead-only
Read recorded metrics for an account (counters + recent events). Read-only, never starts a browser.
Inputs: account (required). Returns: result
get_notificationsBRead-only
Read bounded visible X notifications from a verified All or Mentions tab.
Reports observed actors, related posts, type evidence and unread state; absent evidence remains unknown. Visiting this view may mark X notifications read. Results cover only the visible supported UI.
Inputs: account (required), view, limit. Returns: object
get_profileCRead-only
Read public profile details using visible X page elements.
Inputs: account (required), profile (required). Returns: object
get_profile_connectionsBRead-only
Read visible public followers/following. This is not a private address book or complete contact export.
Inputs: account (required), profile (required), limit, relationship. Returns: object
get_profile_contextBRead-only
Preferred outreach read: exact profile details and bounded authored posts together. No draft or send.
Inputs: account (required), profile (required), post_limit. Returns: object
get_profile_postsBRead-only
Read one author's posts, posts-and-replies, or media tab. feed: posts|replies|media. Results are bounded and partial.
Inputs: account (required), profile (required), feed, limit. Returns: object
get_run_statusARead-only
Poll run_cycle handles. With run_id omitted, lists every run
started since this server booted. Read-only.
Inputs: run_id. Returns: result
get_runtime_infoARead-only
Inspect server/package versions, tool count and configured execution mode, backend workflow support and effective unlock allowance without starting a browser. Contains no credentials, paths, account identities or raw environment values. Configuration does not prove live X authentication; use get_account_health and a bounded read for that.
Inputs: no named parameters. Returns: result
get_session_statusARead-only
Inspect local session lifecycle (cold/starting/ready/busy/closing/ disconnected/closed) without reading cookies or starting a browser. Ready means locally usable transport, not verified live X authentication. Legacy backends may report unknown for a warm session.
Inputs: account. Returns: object
get_threadARead-only
Read bounded visible conversation context around one exact X post. Includes ordered posts and media metadata; parent links remain unknown unless the page proves them. Results are partial, not a full archive. include_images attaches at most five photos/video posters with source mappings for YOUR model to analyze; no server LLM or video transcription. Read this before composing contextual replies or reply threads.
Inputs: tweet_url (required), limit, account, include_images. Returns: object
get_thread_runCRead-only
Read durable local segment states, exact published URLs and next index.
Inputs: run_id (required). Returns: result
get_tweetARead-only
Read-only fetch of one tweet: text, author, public counts, typed media (photos + video posters) and the account's persona. With include_images=true (default) photos also attach as image content so a vision-capable client sees them; the envelope always carries URLs
alt text as the fallback. Nothing is posted; draft mode N/A.
Inputs: account (required), tweet_url (required), include_images. Returns: object
list_accountsARead-only
List configured accounts with secrets stripped (no cookies, no passwords, proxy credentials masked). Read-only, never starts a browser.
Inputs: no named parameters. Returns: result
list_campaignsBRead-only
List this account's local campaigns (without membership details). Status: draft, ready, paused, completed. limit 1-100, with offset.
Inputs: account (required), limit, offset, status. Returns: result
list_draftsARead-only
List drafts, newest first. status filters to one of
pending/approved/partial/executed/failed/rejected/uncertain; account filters by
account id; limit caps the response (max 100). Read-only.
Inputs: limit, status, account. Returns: result
list_leadsARead-only
List this account's local leads, optionally filtering by status or exact tag. limit 1-100; offset supports pagination. No browser starts.
Inputs: account (required), tag, limit, offset, +1 more. Returns: result
list_proxiesARead-only
List proxy pools (size, masked members, rotation cursor), the global pool strategy, each account's assignment, and the global fallback proxy. Read-only, never starts a browser.
Inputs: no named parameters. Returns: result
list_queueARead-only
List queued actions with their full payloads (exactly what will fire), plus per-status counts. limit 1-100; offset pages through local history. Read-only, never starts a browser.
Inputs: limit, offset, status, account. Returns: result
manage_conversationA
Draft one inbox action: accept a request, mute/unmute notifications, or delete a conversation from your inbox. Delete cannot be undone locally and does not promise deletion for recipients. Always requires approval; only verified visible controls are supported.
Inputs: account (required), operation (required), conversation_id (required). Returns: object
opt_out_leadBIdempotent
Permanently suppress outreach to this account's lead handle. This blocks prepared campaign deliveries as well as future drafts.
Inputs: account (required), lead_id (required). Returns: result
pause_account_actionsAIdempotent
Pause browser actions durably until the operator explicitly resumes them.
Inputs: account (required). Returns: object
post_tweetB
Post a tweet from an account. media = local file paths; community
= target community id; quote_tweet_url = post it as a quote of that
tweet, with the text and media in the quote (async browser backend).
In draft mode (default) returns a draft for review and posts nothing;
approve_draft(draft_id) executes it.
Inputs: text (required), account (required), media, community, +1 more. Returns: result
prepare_campaign_messagesA
Prepare at most 20 personalized private-message drafts for eligible new/qualified leads on an active account. Skip opt-outs and prior pending/delivered members. Preparation marks the campaign ready; review every exact payload and approve each draft individually. No bulk sends, browser, scraping, or LLM call occurs, even when draft_mode is off.
Inputs: account (required), campaign_id (required), max_messages. Returns: result
prepare_outreachA
Fetch profile details and up to 10 authored posts, then prepare ONE local DM draft with your exact supplied text. Use get_profile_context to research and write a personalized message first. Context is a bounded visible subset; it is not a complete post history. No server LLM or message send occurs. Review the returned recipient, context and draft, then approve_draft individually; draft_mode does not bypass review. Opt-outs and account activation are rechecked after the browser read.
Inputs: account (required), profile (required), message_text (required), post_limit. Returns: object
prepare_replyARead-only
Read-only reply context: the tweet's text AND media (photos attach as image content so your model can see them), plus the account's keywords and persona. YOUR model writes the reply itself, no server-side LLM, nothing posted. Then pass your text to reply_to_tweet(text=...) or queue_engagement(action="reply", text=...).
Inputs: account (required), tweet_url (required), include_images. Returns: object
prepare_threadA
Stage 1-20 exact text/media posts for review; always creates a draft. Optional reply_to starts a reply thread. Every later post replies to the immediately preceding confirmed new post. No LLM is called.
Inputs: posts (required), account (required), reply_to. Returns: result
preview_campaign_messageARead-only
Preview one campaign member's exact personalized message and local eligibility without creating drafts, reserving a lead, or sending. Includes pause, opt-out and prior-reservation reasons. Eligibility is a snapshot; preparation and approval recheck it. No browser or LLM.
Inputs: account (required), lead_id (required), campaign_id (required). Returns: result
process_queueA
Drain queued actions with jittered pacing and daily caps. THIS
CALL IS THE APPROVAL GATE for queued work: items execute through the
same pacing/dedup/metrics path as every other write action. With
account omitted, every account with due items drains (each bounded
by max_actions and its own daily caps).
Inputs: account, max_actions. Returns: result
queue_engagementA
Queue a like, retweet, or reply for paced execution. action is
one of like/retweet/reply. reply requires text ("auto" scrapes the
tweet and generates the reply NOW, so the stored payload is final).
Nothing executes until process_queue is called.
Inputs: action (required), account (required), tweet_url (required), text. Returns: result
queue_postA
Schedule a post for paced execution. Pass text verbatim, or a
topic to generate the text NOW with the configured LLM (the stored
payload is final, list_queue shows exactly what will be posted).
not_before is an optional ISO 8601 timestamp. Nothing executes
until process_queue is called (or auto_drain, if enabled).
Inputs: account (required), text, media, topic, +2 more. Returns: result
reject_draftAIdempotent
Reject a pending draft so it can never be approved. Local status change only, nothing touches X.
Inputs: draft_id (required). Returns: result
remove_accountADestructiveIdempotent
Remove an account from config/accounts.json. Requires confirm=true. Closes the account's warm session. The backup written during the mutation (config/backups/) is the recovery path; the cookie file is left on disk.
Inputs: account (required), confirm. Returns: result
remove_bookmarkB
Prepare removal of one bookmark. Always requires approve_draft.
Inputs: account (required), tweet_url (required). Returns: object
remove_proxyADestructiveIdempotent
Remove a proxy URL from a pool. Requires confirm=true. The response lists accounts still assigned pool:. Emptying a referenced pool is allowed but reported (those accounts fail closed until reassigned).
Inputs: pool (required), proxy_url (required), confirm. Returns: result
reply_to_tweetA
Reply to a tweet. Pass explicit text, or "auto" to generate the
reply with the LLM from the tweet's actual content (a read-only fetch
is performed so the draft shows the real reply text). In draft mode
(default) nothing is posted until approve_draft(draft_id). media accepts
local attachments on the async browser backend. For conversation-wide
context, read get_thread before supplying explicit reply text.
Inputs: account (required), tweet_url (required), text, media. Returns: result
research_and_stageA
Search each keyword, keep relevant tweets (keyless heuristic), write a reply per tweet with the server-side LLM (persona-aware), and stage one reply DRAFT per tweet. Nothing is posted, review with list_drafts and approve selectively. Requires the 'llm' block in settings.
Inputs: account (required), keywords (required), max_items. Returns: result
resolve_action_outcomeAIdempotent
After inspecting X, record succeeded or not_sent for an uncertain action. Repairs associated draft/queue history; queued not_sent work is cancelled and never resent automatically. Incorrect not_sent can cause duplicates.
Inputs: account (required), action_id (required), observed_outcome (required). Returns: object
resume_account_actionsAIdempotent
After resolving a challenge, probe the signed-in home page and clear the local pause. Does not reset budgets.
Inputs: account (required). Returns: object
run_cycleA
Run the legacy batch automation cycle in the background and return
a run handle immediately (progress goes to the logs, this never
silent-blocks). account limits the run to one account; pipelines
is a comma-separated subset of: competitor_reposts, keyword_replies,
keyword_retweets, likes, content_curation, community_engagement
(mapped onto the account's ActionConfig enable flags for this run
only, config files are never mutated; same names as the CLI's
x-use run --pipeline). Draft mode does not apply to batch cycles.
Inputs: account, pipelines. Returns: result
search_conversationsBRead-only
Search the visible inbox locally; results are partial, not full message-history search.
Inputs: query (required), account (required), limit, folder, +2 more. Returns: object
search_profileARead-only
Read recent posts from ONE X profile. profile takes a handle
("@nasa" or "nasa"), a profile URL, or a tweet URL (which resolves to
its author). Read-only (draft mode does not apply, nothing is
posted), but it reuses the account's browser session. account
defaults to the first active configured account.
Use this to watch specific people and competitors, and to re-read one
of your own published posts for its public counts; use search_tweets
for topic and keyword discovery. Profile timelines include pinned
posts and reposts, so check user_handle before treating a result as
the profile owner's own writing.
include_images=true additionally attaches the first photo of up to
5 posts as image content (bounded); the per-post media URLs + alt
text are always present.
Inputs: profile (required), limit, account, include_images. Returns: object
search_tweetsARead-only
Search recent X posts for a query string. Read-only (draft mode
does not apply). Reuses the account's warm browser session. account
defaults to the first active configured account.
include_images=true additionally attaches the first photo of up to
5 tweets as image content (bounded); the per-tweet media URLs +
alt text are always present.
Inputs: keywords (required), limit, account, include_images. Returns: object
send_messageB
Prepare a reviewed DM draft for one handle or conversation. Always requires approve_draft, even when draft_mode is off.
Inputs: text (required), account (required), recipient (required). Returns: object
set_account_activeAIdempotent
Enable or pause an account without deleting it. Pausing closes its warm browser session.
Inputs: active (required), account (required). Returns: result
set_campaign_statusAIdempotent
Set local campaign status: draft, ready, paused, completed. Pausing immediately blocks preparation and approval of existing drafts. Resume a paused campaign with ready. completed is terminal.
Inputs: status (required), account (required), campaign_id (required). Returns: result
test_proxyARead-only
Probe the effective proxy in an isolated, cookieless browser using the configured backend; visits only an IP-echo endpoint, never X. Explicit proxy_url wins; account assignments inherit the global proxy. Returns validated egress IP, startup/request latency and masked route. Async HTTP(S) authentication is supported; legacy Selenium rejects authenticated proxies. Legacy round_robin probes advance its cursor.
Inputs: account, proxy_url. Returns: result
undo_retweetA
Prepare removal of your repost of one exact post. Always requires approve_draft.
Inputs: account (required), tweet_url (required). Returns: object
unfollow_profileA
Prepare unfollowing one exact X profile. Always requires approve_draft.
Inputs: account (required), profile (required). Returns: object
unlike_tweetA
Prepare removal of your like from one exact post. Always requires approve_draft.
Inputs: account (required), tweet_url (required). Returns: object
unlock_inboxAIdempotent
Unlock the owner's encrypted inbox once using a server environment variable. PIN never appears in tool arguments, drafts, logs or stored state. Inspect get_account_safety for remaining attempts and reset time; never loop on a rejected PIN or challenge.
Inputs: account (required), pin_env_var. Returns: object
update_accountAIdempotent
Partially update an account. Only provided fields change (None =
leave unchanged); pass an empty string to clear proxy. persona
sets the freeform persona text; pass an empty string to clear it.
cookie_file re-imports cookies like add_account. The account's
warm browser session is closed, so the next action cold-starts on
the new config.
Inputs: account (required), proxy, persona, is_active, +5 more. Returns: result
update_lead_statusAIdempotent
Change a lead's local status. Only new/qualified leads can receive campaign drafts. opted_out permanently suppresses further outreach.
Inputs: status (required), account (required), lead_id (required). Returns: result
upsert_leadAIdempotent
Add/update one local lead by X handle (unique per account). Omitted fields retain their values; notes stay local and never enter messages. Status: new, qualified, contacted, replied, closed, opted_out. Opt-out suppression is permanent and cannot be removed by upserting.
Inputs: handle (required), account (required), tags, notes, +3 more. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 82 tools
Several clusters are hard to separate: get_profile/get_profile_posts/get_profile_context/search_profile all read profile content, get_thread/get_conversation/search_conversations overlap on reading context, and post_tweet/generate_and_post/queue_post/prepare_thread/draft_post_variations all produce posts through different paths. Likewise reply_to_tweet vs prepare_reply vs queue_engagement vs engage, and approve_draft vs process_queue vs run_cycle all represent competing write pipelines. Descriptions are verbose and do try to differentiate them, but the agent must read long caveats to pick correctly, so boundaries are not self-evident.
Nearly everything follows a snake_case verb_noun convention (get_*, list_*, create_*, add_*, remove_*, prepare_*, approve_*, cancel_*), which makes the surface predictable. Minor deviations exist: bare verbs like engage and run_cycle, and noun-leading names like draft_post_variations and research_and_stage.
82 tools is far beyond a well-scoped set; the server covers accounts, proxies, drafts, queueing, campaigns, threads, conversations, analytics and safety, so the breadth is real, but many tools are narrow variants (bookmark_tweet vs remove_bookmark, unlike_tweet vs undo_retweet). The count makes discovery and correct selection costly even though each tool is individually documented.
Coverage is unusually thorough: full account/proxy lifecycle, lead and campaign management, draft/queue/approval flow, thread runs, conversations, bookmarks, analytics and safety recovery. Remaining gaps are minor (e.g. no lead deletion, no campaign creation-vs-edit path beyond status, limited inbox message-history search), and workarounds mostly exist.
Maintenance
Related MCP Connectors
Post & schedule to 32+ networks: X, Facebook, Instagram, Threads, LinkedIn, TikTok, YouTube, Reddit.
Your agent posts, schedules, queues and threads posts on X, with a human approval queue.
Draft, preview, schedule and publish posts to X, LinkedIn, Bluesky and 5 more social networks.
Live X/Twitter data: profiles, tweets, search, followers, lists and trends. 29 read-only tools.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceProvides programmatic X (Twitter) engagement via MCP, offering 24 tools for search, timelines, notifications, bookmarks, profiles, and tweet actions through a headless browser.MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to fully automate X/Twitter interactions including searching, posting, engaging, and managing multiple accounts without paid API keys.MIT
- AlicenseAqualityDmaintenanceEnables Twitter/X marketing automation, including tweet search, account analysis, trending topics, and optional AI-generated tweets and scheduling.6MIT

OpenTweet MCP Serverofficial
AlicenseAqualityCmaintenanceProvides 30 tools to post, schedule, thread, and analyze X (Twitter) content, including AI media generation and multi-account management. No X developer account or OAuth setup required.30191 npmMIT