Skip to content
LinkedCamp

Trust center

Trust center

Last updated:

How LinkedCamp keeps client LinkedIn accounts, client data and AI-assistant access under your control, with links to every policy.

Client accounts

How LinkedCamp keeps client accounts safe

Every client account is separated from every other, on its own IP, with limits set for that account.

A dedicated IP per account

A dedicated, country-matched IP per account on every plan. No IP sharing across customers, so one customer's activity doesn't affect another's account.

Client isolation, enforced on the server

Each client is a separate sub-account with its own campaigns, inbox, mailboxes, dedicated IP and API key. The MCP server checks every sub-account id against your roster and refuses anything else.

Smart limits and warm-up

Smart limits that adapt to account age and acceptance rate, and automatic warm-up, for each client account separately.

AI & MCP access

What your AI assistant can and can't do

Your assistant acts with your permissions and nothing more. Read-only and write tools run when you ask; the destructive ones (launching a campaign, sending a message, publishing a post, deleting a record) ask you first, every time. Billing, team roles, stored credentials and LinkedIn account connection are not exposed to the assistant, and blocked calls return a 403.

Sending a message is a destructive tool: your assistant must ask, and you must confirm, every time.

Tools follow your plan and team role; hidden tools don't appear in tools/list. LinkedCamp receives only the tool calls your AI client makes and their inputs, not the rest of the conversation, which stays with your AI provider.

MCP tool permissions

133 tools
Read-only
64
Looks things up
  • analytics_campaign
  • conversation_list
  • linkedin_account_list
Change data
69
Includes the 28 destructive
  • campaign_create
  • lead_note_add
  • lead_update
Destructive
28
Of the 69 · ask first
  • campaign_launch
  • message_send
  • campaign_delete
Never available through MCP
  • Billing and payments
  • Agency admin and impersonation
  • White-label config
  • Team invites and roles
  • Reading stored credentials
  • Connecting a LinkedIn account or entering PINs
  • CRM connection setup
  • Rotating the API key

AI agents

AI agents start in review-first mode

Built-in agents start in a 7-day preview where you approve every message. AI actions that reach prospects wait for approval.

  • Bring your own key: agents can run on your own API key from OpenAI, Anthropic, Google Gemini or DeepSeek, so agent requests go to the provider you choose.

How agentic automation works →

Data and privacy

Where your data rules are written down

Retention, legal bases and your rights are in the Privacy Policy. A standard Data Processing Agreement is available on request from support@linkedcamp.com.

LinkedIn's rules

Lower risk, not zero risk

Any automation carries some risk of restriction, so it's used at your own risk. LinkedCamp reduces that risk with a dedicated IP per account, smart limits and warm-up, but can't remove it. Keep each client's volumes modest and messages relevant.

LinkedCamp is not affiliated with or endorsed by LinkedIn.

Documents

Policies, docs and contacts

Run client outreach with the guardrails on.

Start a 14-day free trial. Every client account gets its own dedicated IP, and your assistant asks before it launches, sends, publishes or deletes anything.