MCP server for interacting with your LINE Official Account
Nothing to run from here
Wazap is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
94/100
trust score
100
downloads a day
local
runs on your machine
6k
downloads in 30d
What it does
WhatsApp for your agent: chats, messages, media, contacts and groups behind MCP tools.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user wazap-mcp '--' npx -y wazap-mcp
# 2 — ask your agent for something these tools do
# learn, get_status, link_account and 19 more
Wazap costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publisher
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Wazap as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user wazap-mcp '--' npx -y wazap-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
Nothing is required, but it can take WAZAP_READ_ONLY (Refuse every write. The write tools are not registered at all.), WAZAP_DATA_DIR (Where wazap keeps credentials, history and media. Defaults to ~/.wazap.) if you want to set them.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
22 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
The guide to these tools: ids, account_id with several accounts, the workflows, the message shape and what to do about each error code. Read it once, before the other tools; it never touches WhatsApp.
No parameter schema on file.
Whether the account works: status (connected, not_linked, linking with its pairing code…), sync, how fresh the history is, webhook delivery, versions. accounts lists every account, with default. Call it on NOT_CONNECTED, NOT_LINKED or SYNC_IN_PROGRESS.
No parameter schema on file.
Pair wazap with the user's WhatsApp when get_status says not_linked, logged_out, session_corrupt or auth_failure. Show the user the code it returns: WhatsApp → Settings → Linked devices → Link a device → Link with phone number instead. Then poll get_status every 10 s until connected.
No parameter schema on file.
Conversations, most recently active first: chat_id, name, type, unread count, last message, and whether archived, pinned, muted or left. Use it to find a chat_id.
No parameter schema on file.
Read one chat, oldest to newest: the latest messages, or older ones with before (wazap asks the phone when the local history runs out). chat_id "status" reads the stories people posted, newest first, from the last hours (at most a day); they show nowhere else.
No parameter schema on file.
Keep what the user says about someone, locally, never on WhatsApp: a note, tags, details find_contact matches ({"relatie": "mama"}), or handled: true for an ask dealt with elsewhere. #private keeps their words out of what you did not ask about them by name; #no-catchup keeps them out of catch_up.
No parameter schema on file.
Wait for messages from other people, up to timeout_seconds, and return them with a cursor: pass it to the next call so nothing that lands between calls is missed. addressed_to_me wakes only for direct messages, mentions and replies to the user. For agents that stay on the line.
No parameter schema on file.
Find messages by meaning and by words at once, in every chat or one: a paraphrase still hits; another language seldom does, so ask in the chat's language. match: "words" for exact words (a number, a URL). since, until and from narrow it; coverage and freshness say how much history was searched.
No parameter schema on file.
One message in full: the message it quotes, each reaction with who left it, poll votes and event answers by person, its media, and on the user's own messages who it reached and who read it. Without account_id the id is looked for on every account.
No parameter schema on file.
Who a name, nickname, relationship ("mama"), group, number or id means, before drafting. resolved: contact.chat_id, with number, note, tags, details and, in a write session, context. ambiguous or not_found: ask the user; never send to a guess. A name on several accounts: see fix. tag: its people.
No parameter schema on file.
The phone's address book on one account, a page at a time: saved name, phone (E.164), chat_id, contact_id, and when the last direct message was and its direction. No message words; #private people left out. address_book_synced false: not arrived yet.
No parameter schema on file.
A group's name, description, owner and participants (up to 500), whether the account is admin, its settings (info_locked, member_add_mode, join_approval, disappearing_seconds), its community, and the invite link for an admin. Call it before manage_group.
No parameter schema on file.
What a message's media holds: a voice note or audio as its transcript (kept once made; an API provider bills it), or with save_to its file; a photo attached as an image; any file saved at path on the machine running wazap; a copy kept on arrival answers first. MEDIA_UNAVAILABLE: gone, none kept.
No parameter schema on file.
Drafts text, media, polls, locations, forwards; sends nothing. Call it as soon as you have recipient and text: it returns draft_id and preview (recipient, number, exact text). Show that preview; confirm_send only on a yes to this text and recipient — a send in the same request is that yes.
No parameter schema on file.
Replace the text of a message the linked account sent. WhatsApp allows it for 15 minutes after sending; later, send a correction instead.
No parameter schema on file.
React to a message with one emoji, or pass "" to take your reaction off.
No parameter schema on file.
Send a draft the user approved — this text, this recipient: the only call that sends, once per draft. A yes about something else: show the preview again and ask. Expired, missing or stale: draft again, show the new preview, ask again. SEND_OUTCOME_UNKNOWN: read_messages, never redo it unasked.
No parameter schema on file.
Delete a message; confirm with the user first. for_everyone: true retracts it for everyone within 2 days: the account's own message, or anyone's in a group where it is admin. for_everyone: false removes any message from the linked account's devices only.
No parameter schema on file.
Change a chat on WhatsApp at once, with no draft: archive, pin, mute, mark_read or mark_unread; pin_message (for everyone) or star_message on a message_id; clear or delete it for the linked account; block a person. Say what will change and wait for a yes.
No parameter schema on file.
Create a group, join one from an invite (preview first, then confirm: true), or administer one: members, name, description, photo, invite link, join requests, settings, leave. Most actions need admin. Every change shows to all members at once: say what will change, wait for a yes.
No parameter schema on file.
The account's scheduled messages: list (waiting ones soonest first, then the week's sent, missed, failed and cancelled ones) or cancel one by schedule_id before it goes. To schedule one, send_message with send_at, then confirm_send. Tell the user about every missed or failed one.
No parameter schema on file.
The user's reminders about a chat, kept locally and posted to their webhook when due: create (chat_id, note, after_hours or due_at); condition "no_reply" fires only if nobody wrote there since ("if Ana has not answered in 2 days"). list, cancel. Nothing goes to WhatsApp.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Wazap or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 94 / 100
Release 1.4.0 · scanned 14 Oct 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score94
Findings9 threat · 0 capability
Capabilityhigh
Coveragesource
Verificationsource
Surface digest9c14360cb5b690aa5b0cb010aa93415dd512b4ae0946ae0a84d8a5febc820dae
Score vectorthreat 100 → client 94
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfacenot available
Implementation sourceread
Package metadataread
Live transportnot available
No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| No per-tool rows for this release — see the note below. | ||
can send data / act on an external service
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest9c14360cb5b690aa
Scanned2026-10-14
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 1.4.0. Pin a release below and the command asks for exactly that one.
1.4.0
11 Oct 2026
latestNo release note was published with this version.
1.2.0
6 Oct 2026
No release note was published with this version.
1.1.0
28 Sep 2026
No release note was published with this version.
1.0.3
23 Sep 2026
No release note was published with this version.
1.0.2
21 Sep 2026
No release note was published with this version.
0.23.1
17 Sep 2026
No release note was published with this version.
0.20.2
15 Sep 2026
No release note was published with this version.
0.18.4
13 Sep 2026
No release note was published with this version.
0.13.0
5 Sep 2026
No release note was published with this version.
0.10.0
undated
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Wazap can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.