TeleBoost

npx add-mcp 'https://mcp.teleboost.app/mcp'

Getting started

The TeleBoost API is available on the Pro and Business plans. Create a key under Settings → Developers → API keys — it's shown once, so copy it then. All requests use HTTPS and return JSON.

curl https://teleboost.app/api/v1/contacts \
  -H "Authorization: Bearer tb_live_your_key_here"

List endpoints return { "data": [...], "next_cursor": "...", "has_more": true }. Pass next_cursor back as ?cursor= to page.

Authentication & scopes

Two mechanisms, one scope model. Use an API key (Authorization: Bearer tb_live_…) for scripts and automation, or OAuth 2.1 for AI apps connecting over MCP. Keys are workspace-scoped: a key is bound to your solo data or one team at creation.

Scope	Grants
workspace.read	Workspace metadata, team members, plan usage and quotas
accounts.write	Update connected account settings such as daily sending quotas
contacts.read / write	Leads, custom fields, notes, statuses and engagement scores
lists.read / write	Lists and their members
groups.read / write	Saved Telegram groups (campaign & scraping targets)
inbox.read	Conversations and message history across accounts
messages.send	Send messages from a connected account
campaigns.read / write	DM & group campaigns, stats, start / pause
templates.read / write	Message templates and AI personalities
tickets.read / write	Tickets, statuses, priorities, SLA and reminders
analytics.read	Aggregated metrics, dashboards and heatmaps

API resources

All paths are relative to https://teleboost.app/api/v1. A selection is shown below — the full, always-current reference of every endpoint is the OpenAPI spec.

Method	Path	Description	Scope
GET	/accounts	List connected accounts (status only)	workspace.read
PATCH	/accounts/{id}/quota	Update an account's sending quota	accounts.write
GET	/contacts	List and search contacts	contacts.read
POST	/contacts	Create a contact	contacts.write
PATCH	/contacts/{id}	Update status, custom fields	contacts.write
GET	/lists	Browse lists with counts	lists.read
POST	/lists/{id}/contacts	Add contacts to a list	lists.write
GET	/groups	List saved Telegram groups	groups.read
GET	/conversations	List inbox conversations	inbox.read
GET	/conversations/{id}/messages	Message history	inbox.read
POST	/messages	Send a message	messages.send
GET	/campaigns/dm	List DM campaigns	campaigns.read
POST	/campaigns/dm/{id}/start	Start / resume a DM campaign	campaigns.write
POST	/campaigns/group	Create a group campaign	campaigns.write
GET	/templates	List message templates	templates.read
GET	/personalities	List AI personalities	templates.read
GET	/tickets	List tickets	tickets.read
POST	/tickets	Create a ticket	tickets.write
POST	/reminders	Create a reminder	tickets.write
GET	/analytics/summary	Aggregated KPIs	analytics.read
GET	/analytics/activity	7×24 inbound activity heatmap	analytics.read
GET	/workspace/usage	Plan usage and API quotas	workspace.read

MCP server

TeleBoost runs a hosted MCP server so any MCP-compatible AI — Claude, ChatGPT, Cursor, Zed, and automation tools like n8n and Make — can work with your pipeline directly.

MCP endpoint:  https://mcp.teleboost.app/mcp

1. Copy the URL from Settings → Developers → Connected AI apps
2. Add it as a custom connector in your AI tool
3. Authorize via the TeleBoost consent screen (OAuth 2.1 + PKCE)

The server exposes tools that mirror the API: search_contacts, get_conversation, create_ticket, send_message, start_campaign, get_analytics_summary, and more. Tools outside a connection's granted scopes are never exposed.

Webhooks

Register an HTTPS endpoint under Settings → Developers → Webhooks. Each delivery carries an X-TeleBoost-Signature header — an HMAC-SHA256 of the body with your endpoint secret. Verify it before trusting the payload.

import hmac, hashlib

def verify(body: bytes, header: str, secret: str) -> bool:
    expected = "sha256=" + hmac.new(secret.encode(), body, hashlib.sha256).hexdigest()
    return hmac.compare_digest(expected, header)

Deliveries retry with exponential backoff for up to 24 hours; endpoints that keep failing are auto-disabled with an alert. Available events:

Rate limits

Limits are per plan and returned on every response via X-RateLimit-* headers. A 429 includes Retry-After.

Plan	Per minute	Per day	Sends / hour
Pro	120	10,000	30
Business	300	50,000	100

Message sends also respect each account's own safety quota — the hourly API cap is in addition to, not instead of, those limits.