Skip to content

Teppek MCP server

com.teppek/teppekteppek.comstreamable-httpv1.3.1Scanned

Agent-accessible map of global supply & demand: jobs, homes, cars & services your AI can act on

16 tools: 5 can delete or cancel things, 1 do something that can't be undone, 4 change things and 6 only read.

Tools: 5 delete, 1 irreversible, 4 write, 6 read.
  • Reads private data · 2
  • Reads text others write · 8
  • Can send data out · 2
  • 1 finding

Tools listed without sign-in. Remote posture

With Anlyon

5 tools here can delete or cancel things and 1 does something that can't be undone. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Teppek MCP server can do

The Teppek MCP server is published by teppek.com, a domain the MCP Registry verified. It has 16 tools. It can delete or cancel things (5 tools: send_message, update_listing, delete_listing, reply_to_conversation and 1 more), do things that cannot be undone, such as sending, paying or publishing (create_listing) and create or change data (4 tools: request_image_upload, get_conversation, set_listing_status and renew_listing). 6 tools only read.

2 tools read private data (list_conversations and get_conversation); 8 tools bring in text other people can write (search_listings, semantic_search, get_listing and 5 more); 2 tools can send data out of your system (send_message and reply_to_conversation). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 1 finding (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in, and its tools changed 3 times since Anlyon Radar first saw them.

Last scanned Nov 16, 2026, version 1.3.1.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

16 tools

Deletes or cancels 5Claude Code: deny
  • delete_listingDelete (soft-delete) one of the USER'S OWN listings by listing_id: it leaves the marketplace and cannot be restored through these tools.No findings

    Why it is in this group

    • The name "delete_listing" says "delete": a word for removing or ending something.
    • The description says "cannot be restored".
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Delete (soft-delete) one of the USER'S OWN listings by listing_id: it leaves the marketplace and cannot be restored through these tools. Use it ONLY when the user clearly asks to delete the listing. If they say "take it down", "pause it" or "I sold it", use set_listing_status (hidden or sold) instead — that is reversible. Ask for confirmation when the request is ambiguous. Only the owner can delete. No credit is refunded — the listing was already published. Returns { success: true } when the listing was deleted.

    Annotations the server declares

    • title: Delete my listing
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • reply_to_conversationReply in an EXISTING message thread by thread_id — the message goes to the other participant.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "reply_to_conversation" says "reply": a word for sending, and what is sent cannot be taken back.
    • The description says "cannot be undone".
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "reply_to_conversation" says "reply": it sends or publishes something to other people.
    Description, as the server wrote it
    Reply in an EXISTING message thread by thread_id — the message goes to the other participant. Use this to follow up in a conversation that already exists, whether the user is the listing owner answering an inquiry or the inquirer following up — e.g. "tell the buyer the car is still available" or "ask the landlord if I can view it on Saturday". For the FIRST message to a listing found in search (you have a listing_id but no thread yet), use send_message instead. Contact routes through Teppek; raw personal details are never exposed. Find thread_id via list_conversations / get_conversation. Returns { thread, message }: the message just sent (id, body, created_at) and the updated thread. Requires the user to be connected to Teppek (authenticated); the body limit is 2000 characters, and a sent message cannot be undone.

    Annotations the server declares

    • title: Reply in my conversation
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • send_messageSTART contact with the owner of a Teppek LISTING on the user's behalf (identified by the listing) — the primary FIRST-CONTACT path (there is no price-offer/accept/reject flow).Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "send_message" says "send": a word for sending, and what is sent cannot be taken back.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "send_message" says "send": it sends or publishes something to other people.
    Description, as the server wrote it
    START contact with the owner of a Teppek LISTING on the user's behalf (identified by the listing) — the primary FIRST-CONTACT path (there is no price-offer/accept/reject flow). Opens a private thread with the owner; contact routes through Teppek and never exposes raw personal details. To continue a thread that already EXISTS (you have a thread_id from list_conversations), use reply_to_conversation instead. External-feed listings (e.g. imported jobs) have no Teppek owner, so this returns a redirect to the external application URL instead of creating a thread. Requires the user to be connected to Teppek: an anonymous call returns an AUTH_REQUIRED error with an authorize_url — surface 'Connect with Teppek' to the user, then retry.

    Annotations the server declares

    • title: Message a listing owner
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • update_listingEdit fields (title, description, price, currency, location, details) on one of the USER'S OWN listings, identified by listing_id.No findings

    Why it is in this group

    • The name "update_listing" says "update": a word for a change.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Edit fields (title, description, price, currency, location, details) on one of the USER'S OWN listings, identified by listing_id. Use it when the user wants to correct or change an existing listing — "drop the price to 900", "fix the typo in the title", "the flat now has 3 rooms". Do not use it for photos (update_listing_images), for marking sold or hiding (set_listing_status), for extending the expiry (renew_listing), or to change vertical/intent_type/role — those are fixed, so publish a new listing instead. Only the owner can edit. Editing is free (no credit charge). Price/currency-only edits stay live; changing the title, description, location or details re-triggers AI moderation: the listing returns to 'pending' and is offline until cleared — tell the user this before editing a live listing. Returns { success, listing } with the updated listing and its new status.

    Annotations the server declares

    • title: Edit my listing
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • update_listing_imagesAdd and/or remove photos on one of the USER'S OWN existing listings (max 6 total).No findings

    Why it is in this group

    • The name "update_listing_images" says "update": a word for a change.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Add and/or remove photos on one of the USER'S OWN existing listings (max 6 total). Use it when the user wants to add, replace or remove photos on a listing that already exists — to replace a photo, remove the old key and add the new one in the same call. Do not use it while publishing a new listing (pass images to create_listing instead) or to change text fields (update_listing). First call request_image_upload for each new image and upload the bytes, then pass the returned keys in `add`; a key with no uploaded file fails with image_not_uploaded. Adding any image re-triggers AI moderation — the listing returns to 'pending' until the new images are cleared. Removing images leaves a live listing live; on a pending or rejected listing it requests a new moderation verdict. Free (no credit). Returns { success, listing } with the updated listing and its status.

    Annotations the server declares

    • title: Update my listing photos
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: false
Can't be undone 1Claude Code: ask
  • create_listingPublish a NEW listing on behalf of the user across any of the 4 verticals and 12 roles.Reads text others writeNo findings

    Why it is in this group

    • The name "create_listing" says "create": a word for a change.
    • The description's first verb is "publish": a word for a change that another call does not undo.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Publish a NEW listing on behalf of the user across any of the 4 verticals and 12 roles. Use it only when the user clearly asks to post something — sell or rent out an item or home, advertise a job, publish a CV, offer or request a service. Do not use it to change an existing listing (update_listing, update_listing_images, set_listing_status) or to look for listings (search_listings). Recommended order: get_entitlements → resolve the address to coordinates → optionally request_image_upload and upload each photo → create_listing. `location` must be coordinates ({ lat, lon }) — a free-text address is not accepted, so geocode it first. Publishing needs a free listing slot and may consume credits (included listings cost none); get_entitlements reports the slots, credit cost and balance. A reached limit makes this return an ACTIVE_LISTING_LIMIT_REACHED error carrying a pay_url. Held for AI moderation (status starts "pending"); any credits spent are refunded if rejected, EXCEPT for prohibited or spam content, which forfeits the credit. To attach photos, first call request_image_upload for each image, upload the bytes, and pass the returned keys in `images` — a key with no uploaded file fails with image_not_uploaded, and every image is AI-moderated with the text before going live. Requires the user to be connected to Teppek: an anonymous call returns an AUTH_REQUIRED error with an authorize_url — surface "Connect with Teppek" to the user, then retry.

    Annotations the server declares

    • title: Publish a listing
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
Changes things 4Claude Code: ask
  • get_conversationRead ONE of the user's OWN message conversations by thread_id, including its messages.Reads private dataReads text others write1 finding

    Why it is in this group

    • The name "get_conversation" says "get": a word for reading.
    • The server marks it as not read-only (readOnlyHint: false).

    What it can reach

    • private-dataThe name "get_conversation" says "get" and "conversation": it reads messages.
    • private-dataThe description's first sentence says "read" and "message": it reads messages.
    • untrusted-inputThe name "get_conversation" says "get" and "conversation": it reads messages, which other people can write.
    • untrusted-inputThe description's first sentence says "read" and "message": it reads messages, which other people can write.

    Findings

    • Warn

      The server marks "get_conversation" as not read-only (readOnlyHint: false), and its words say it reads.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "get_conversation" says "get": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Read ONE of the user's OWN message conversations by thread_id, including its messages. Use it when the user wants to read a specific conversation ("what did the landlord say?") or before writing a reply that depends on what was said. Side effect: incoming messages in the thread are marked read — for a quick inbox overview or unread counts use list_conversations instead, so messages the user has not seen stay unread. Returns { thread, messages: [...] }; each message has id, sender_name, body, created_at, read_at and is_mine (true = sent by the user). Only threads the authenticated user participates in are returned; any other thread_id returns an error. Get the thread_id from list_conversations, then answer with reply_to_conversation.

    Annotations the server declares

    • title: Read a conversation
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • renew_listingUpdate the expiry of one of the USER'S OWN listings (renew it): the listing runs for another 30 days from now (native listings lapse ~30 days after posting).No findings

    Why it is in this group

    • The description's first verb is "update": a word for a change.
    Description, as the server wrote it
    Update the expiry of one of the USER'S OWN listings (renew it): the listing runs for another 30 days from now (native listings lapse ~30 days after posting). Use it when the user wants to keep an ad running ("extend my listing", "my ad expired, put it back") or when list_my_listings shows it as expired. Renewing an expired listing returns it to active. Do not use it on a pending, rejected, sold or draft listing — the call fails: a pending one only needs moderation to finish, a rejected one needs its content changed with update_listing, and a sold one is closed. Free — no credit charge. Returns { success, listing } with the renewed listing.

    Annotations the server declares

    • title: Renew my listing
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • request_image_uploadRequest a short-lived upload grant for ONE image file — step 1 of attaching a photo (any vertical/role).No findings

    Why it is in this group

    • The name "request_image_upload" says "request": a word for a change.
    Description, as the server wrote it
    Request a short-lived upload grant for ONE image file — step 1 of attaching a photo (any vertical/role). Use it when the user wants a photo on a new listing (then create_listing) or added to an existing one (then update_listing_images) — call it once per image, maximum 6 images per listing. Do not use it when you only have an image URL: it needs the raw file bytes (JPEG, PNG or WebP, up to 10 MB), and it attaches nothing by itself. Returns { key, upload_url, upload_token, headers_required }: PUT the raw image bytes to upload_url with exactly those headers, then pass the returned key in create_listing's `images` array or update_listing_images' `add` array. A key whose bytes were never uploaded fails later with image_not_uploaded. Images are AI-moderated for content safety before the listing goes live. Requires the user to be connected to Teppek (authenticated).

    Annotations the server declares

    • title: Request an image-upload grant
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • set_listing_statusSet the lifecycle status of one of the USER'S OWN listings — the reversible alternative to delete_listing.No findings

    Why it is in this group

    • The name "set_listing_status" says "set": a word for a change.
    Description, as the server wrote it
    Set the lifecycle status of one of the USER'S OWN listings — the reversible alternative to delete_listing. Use status='sold' when the user says the item is sold, the job is filled or the home is let; 'hidden' to take a live listing offline for a while ("pause my ad for a week"); 'active' to bring a hidden listing back. Do not use it to extend an expired listing (use renew_listing), to edit content (update_listing), or to force a pending or rejected listing live: moderation states (pending/rejected) cannot be set, and 'active' only works on a listing that was previously approved and is currently hidden. Free — no credit charge. Returns { success, listing } with the listing's new status.

    Annotations the server declares

    • title: Change my listing status
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Only reads 6Show 6 read-only tools
  • get_entitlementsCheck whether the user can publish right now.No findings

    Why it is in this group

    • The name "get_entitlements" says "get": a word for reading.
    Description, as the server wrote it
    Check whether the user can publish right now. Use it BEFORE create_listing, and whenever the user asks "how many listings can I still post?" or "do I have enough credit?". When you are about to publish, pass the `role` you will use in create_listing and read can_create_for_role — true means go ahead, false means show the user the pay_url instead of attempting the publish; omit `role` for a general overview. Always returns success, membership_key, included (true = listings are covered by the plan), current_active_listings, credit_cost (credits per listing; null when included), wallet_balance, can_pay_from_wallet, can_create_listing and pay_url (set only when the user cannot publish). Paid plans also return max_active_listings and remaining_slots (null = unlimited). The free tier instead caps supply and demand listings SEPARATELY and returns max_active_listings_per_intent, active_supply, active_demand, can_create_supply and can_create_demand — there, can_create_listing is true if EITHER side has room, which is why passing `role` matters: can_create_for_role answers for the side of that role only. Example: role="career_employer" returning can_create_for_role=false with a pay_url means send the user to that link instead of attempting the publish.

    Annotations the server declares

    • title: Check publishing entitlements
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_listingFetch the full detail of ONE Teppek listing by listing_id.Reads text others writeNo findings

    Why it is in this group

    • The name "get_listing" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Fetch the full detail of ONE Teppek listing by listing_id. Use it after search_listings or semantic_search when the user picks a specific result and wants to see, summarize or compare it — e.g. "what is the salary on the second job?", "compare these two flats", "tell me more about that car". Do not use it to browse or count listings (use search_listings); you also do not need it just to show a result list, because search already returns the summary fields. Returns the listing object: id, title, description, price and currency (jobs also carry salary_min/salary_max), vertical, role, status, lat/lon and location_name, vertical-specific `attributes`, images, expires_at, deep_link (the contact or apply route) and the canonical teppek.com `url` the user can open. An unknown or removed id returns an error ("Listing not found") instead of a listing. Read-only and public — no authentication needed.

    Annotations the server declares

    • title: Get listing detail
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • list_conversationsList the user's OWN listing-message conversations — threads they started about someone's listing, and inquiries they received as a listing owner — newest first.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_conversations" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_conversations" says "list" and "conversation": it reads messages.
    • private-dataThe description's first sentence says "list" and "message": it reads messages.
    • untrusted-inputThe name "list_conversations" says "list" and "conversation": it reads messages, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "message": it reads messages, which other people can write.
    Description, as the server wrote it
    List the user's OWN listing-message conversations — threads they started about someone's listing, and inquiries they received as a listing owner — newest first. Resolved from the authenticated agent key — only ever this user's threads. Use it when the user asks "do I have new messages?" or "who replied about my flat?", and before replying, to find the thread_id. It has no side effects (nothing is marked read), so prefer it over get_conversation when you only need an inbox overview or unread counts. Returns { threads: [...] }; each thread has id (the thread_id), listing_id, listing_title, counterpart_name, last_message_preview, last_message_at and viewer_state (unread_count, is_listing_owner, is_last_message_mine). Read a thread's full messages with get_conversation. Page with limit/offset.

    Annotations the server declares

    • title: List my conversations
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_my_listingsList the user's OWN listings (any status), newest first.No findings

    Why it is in this group

    • The name "list_my_listings" says "list": a word for reading.
    Description, as the server wrote it
    List the user's OWN listings (any status), newest first. Resolved from the authenticated agent key — only ever returns this user's listings. Use it when the user asks about their own ads — "how are my listings doing?", "is my car ad live yet?", "why was my listing rejected?" — and to look up the listing_id before update_listing, update_listing_images, set_listing_status, renew_listing or delete_listing. Filter with status: 'pending' = awaiting AI moderation, 'rejected' = declined, 'expired' = lapsed (renew it with renew_listing), 'hidden' = taken offline by the user, 'all' (default) = everything. Do not use it to search other people's listings (use search_listings). Returns { success, listings: [...] }, each with id, title, status, price, currency, vertical and the teppek.com url. Page with limit/offset.

    Annotations the server declares

    • title: List my listings
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • search_listingsFind live jobs, homes/rentals, vehicles, or local services NEAR a place or in a city/country on Teppek — use this for natural requests like "jobs near me", "apartments in Berlin", …Reads text others writeNo findings

    Why it is in this group

    • The name "search_listings" says "search": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Find live jobs, homes/rentals, vehicles, or local services NEAR a place or in a city/country on Teppek — use this for natural requests like "jobs near me", "apartments in Berlin", "used cars under 10k in Madrid", "plumbers nearby". Worldwide, with fresh, location-aware listings refreshed daily. For MEANING-based or fuzzy natural-language intent (not exact keywords or filters), use semantic_search instead. Mechanics: search by vertical (career/real_estate/vehicle/service), role, text, price and a radius around a lat/lon point. The `role` is the perspective you search AS and returns the COUNTERPARTY listings: to find JOB POSTINGS use role="career_seeker" (NOT career_employer, which searches candidate CVs). Always send `role` with text_query: without it the call fails with an error, and vertical alone is not enough. country, category and near work without role; a call with none of text_query, country, category or near returns nothing. Real estate inventory is currently very limited, so an empty real_estate result does not mean the search failed. The response meta.total is the REAL match count (independent of limit); for a multi-word text_query it counts listings matching ANY of the words, so to count a whole occupation/category include its synonyms (e.g. "waiter waitress server"). Use the `country` field for country-scoped totals. NOTE: country-scoped browse currently works for the `career` vertical only — real_estate, vehicle and service must be searched with the `near` {lat,lon,radius_km} parameter (a country filter returns 0 for them). meta.ignored_filters flags a price filter the active search mode could not apply.

    Annotations the server declares

    • title: Search listings
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

32 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 2 outbound tools breaks every one.

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write8
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out2
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__teppek__reply_to_conversation",
      "mcp__teppek__send_message"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:48 UTC, unchanged since Nov 6, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
10
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
https://mcp.teppek.com/.well-known/oauth-protected-resource/mcp
Authorization servers
https://mcp.teppek.com
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
no
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 10 tools that change things (5 destructive, 1 irreversible, 4 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://mcp.teppek.com/mcp: tools/list returned 16 tools to an anonymous client, including send_message (destructive), update_listing (destructive), delete_listing (destructive), reply_to_conversation (destructive), update_listing_images (destructive) and 5 more

    What this check looks for

Posture changes

  1. Nov 6, 2026
    • Legacy HTTP+SSE transport: yes to no

Findings

Matches of fixed checks, with the text they matched.

1 warn
  • The server marks "get_conversation" as not read-only (readOnlyHint: false), and its words say it reads.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "get_conversation" says "get": a word for reading.

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools16 changed.~16 changedView diff
  2. PostureRemote posture changed. Legacy HTTP+SSE transport: yes -> no.View diff
    • Legacy HTTP+SSE transport: yes to no
  3. Tools16 changed.~16 changedView diff
  4. Tools16 changed.~16 changedView diff

Install

Pinned to version 1.3.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http teppek https://mcp.teppek.com/mcp
claude mcp add --transport http teppek https://mcp.teppek.com/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__teppek__search_listings",
      "mcp__teppek__semantic_search",
      "mcp__teppek__get_listing",
      "mcp__teppek__list_my_listings",
      "mcp__teppek__list_conversations",
      "mcp__teppek__get_entitlements"
    ],
    "ask": [
      "mcp__teppek__create_listing",
      "mcp__teppek__request_image_upload",
      "mcp__teppek__get_conversation",
      "mcp__teppek__set_listing_status",
      "mcp__teppek__renew_listing"
    ],
    "deny": [
      "mcp__teppek__send_message",
      "mcp__teppek__update_listing",
      "mcp__teppek__delete_listing",
      "mcp__teppek__reply_to_conversation",
      "mcp__teppek__update_listing_images"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as teppek. Review it before you use it.

Capability tags behind 8 rules
  • mcp__teppek__search_listings untrusted-input
  • mcp__teppek__semantic_search untrusted-input
  • mcp__teppek__get_listing untrusted-input
  • mcp__teppek__send_message untrusted-input, outbound
  • mcp__teppek__create_listing untrusted-input
  • mcp__teppek__list_conversations private-data, untrusted-input
  • mcp__teppek__get_conversation private-data, untrusted-input
  • mcp__teppek__reply_to_conversation untrusted-input, outbound

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.teppek/teppek/verify?hash=471e234b79165c006bd0415a28fc07b947a96ca16e131da18b1c54f402219344" | jq -e '.changed == false'

Pinned hash 471e234b7916, from Nov 16, 2026, 23:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.teppek/teppek/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.teppek/teppek
[![Anlyon Radar](https://radar.anlyon.com/badge/com.teppek/teppek.svg)](https://radar.anlyon.com/servers/com.teppek/teppek)

Source: MCP Registry. Synced Nov 3, 2026, 11:17 UTC.

Teppek MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Teppek MCP server safe?

No scan can promise that. What its definitions show: 16 tools, 10 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Teppek MCP server delete data?

Yes. 5 tools can delete, cancel or revoke something: send_message, update_listing, delete_listing, reply_to_conversation and update_listing_images.

Can the Teppek MCP server do things that cannot be undone?

Yes. 1 tool can send, pay, publish or merge: create_listing.

What tools does the Teppek MCP server have?

16 tools: 5 destructive, 1 irreversible, 4 write, 6 read. send_message, update_listing, delete_listing, reply_to_conversation, update_listing_images, create_listing, request_image_upload, get_conversation and 8 more.

Does the Teppek MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Teppek MCP server send data to outside services?

2 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): send_message and reply_to_conversation.

Has the Teppek MCP server changed its tools?

Yes, 3 times since Anlyon Radar started tracking it, most recently on Nov 16, 2026. Every change is on this page and in its Atom feed.