LinkedCamp for OpenClaw: LinkedIn and email outreach campaigns, prospects, inbox, AI reply agents and analytics through the hosted LinkedCamp MCP server, with approvals before anything reaches people.
Install
openclaw plugins install clawhub:@asveloper/linkedcampLinkedCamp for OpenClaw
LinkedIn automation and multi-channel outreach (LinkedIn + email) for OpenClaw, powered by LinkedCamp.
Run your outreach from any chat OpenClaw is on — WhatsApp, Telegram, Slack, Discord or the Control UI:
- build campaigns, generate LinkedIn and Sales Navigator audience searches, write multi-step sequences, and launch them after you confirm
- review and update leads, notes, tags and blacklists
- work the unified inbox and draft replies
- set up and test AI reply agents and their knowledge bases
- schedule LinkedIn posts
- read campaign, account and funnel analytics, including agency sub-accounts
Your LinkedCamp plan and permissions decide which features are available.
Install
| What you get | Install | |
|---|---|---|
| Plugin (recommended) | LinkedCamp connection, the skill, and approval prompts | openclaw plugins install clawhub:@asveloper/linkedcamp |
| Skill only | The skill. You connect LinkedCamp yourself (below) | openclaw skills install @asveloper/linkedcamp |
Requires OpenClaw 2026.6.1 or newer.
Connect your LinkedCamp account
You need your API key from LinkedCamp → Settings → Integrations. Never paste it into a chat.
Sign in (recommended)
openclaw mcp set linkedcamp '{"url":"https://mcp.linkedcamp.com/mcp","transport":"streamable-http","auth":"oauth"}'
openclaw mcp login linkedcamp
Open the printed link, paste your API key into LinkedCamp's page, and approve.
OpenClaw keeps an encrypted token, not the key. If your Gateway runs on another
machine and the browser can't return to it, finish with
openclaw mcp login linkedcamp --code <code>.
In the Control UI: Settings → MCP → linkedcamp → Sign in.
API key (headless Gateways)
Put the key in the Gateway's environment, for example ~/.openclaw/.env:
LINKEDCAMP_API_KEY=your-api-key
Then reference it. Keep the single quotes so your shell leaves the variable for OpenClaw to fill in:
openclaw mcp set linkedcamp '{"url":"https://mcp.linkedcamp.com/mcp","transport":"streamable-http","headers":{"X-Api-Key":"${LINKEDCAMP_API_KEY}"}}'
One LinkedCamp account per chat member (teams)
On a shared Gateway, each person can connect their own account:
openclaw config set gateway.publicOrigin https://gateway.example.com
openclaw mcp set linkedcamp '{"url":"https://mcp.linkedcamp.com/mcp","transport":"streamable-http","auth":"oauth","oauth":{"identity":"per-requester"}}'
Check it
openclaw mcp probe linkedcamp
Then ask your agent: "Which LinkedCamp account am I connected to?"
Try it
- "Show me my LinkedCamp account and unread inbox counts."
- "Build a draft campaign for heads of customer support at B2B SaaS companies. Show me the audience and the sequence before anything sends."
- "Summarize my campaign performance for the last 30 days."
- "Find my unread conversations, prioritize interested prospects and draft a reply for each. Don't send anything."
Safety
With the plugin installed, the agent stops and asks you to allow once or
deny before anything that reaches people, changes live outreach, or
removes data. You can approve in the Control UI, with /approve, or in a chat
you route approvals to. Each prompt shows what will happen and the details
that matter, such as the message text and the conversation.
You'll be asked before:
- Reaching people: launching or resuming a campaign, sending a message or test email, scheduling or publishing a LinkedIn post, changing posting slots, and turning on an automation.
- Changing live outreach: setting or adding to a campaign's audience, saving or editing a sequence, attaching an AI agent to a campaign, editing an active agent, its knowledge or a persona, overwriting templates, saved replies or notes, and skipping a scheduled step.
- Removing data: deleting campaigns, leads, messages, agents, knowledge, personas, tags, blacklist rules, merge fields, blocked dates or reminders, and blacklisting a lead.
No prompt for reading data, building a draft campaign, adding leads, private draft posts, or agents and automations saved inactive. Saving a sequence asks even on a draft, because it can't tell whether the campaign is live.
The agent also confirms these with you in the conversation first. With the skill alone (no plugin) there are no approval prompts, so that conversation is the only check.
Settings, under plugins.entries.linkedcamp.config:
| Setting | Default | |
|---|---|---|
requireApproval | true | Set false to turn approval prompts off. Not recommended. |
serverName | "linkedcamp" | Only if you connected LinkedCamp under another name. |
To make sure an agent can never call certain tools at all, hide them:
openclaw mcp tools linkedcamp --exclude 'message_send,campaign_launch,*_delete,lead_blacklist'
Not available here: connecting a LinkedIn account (it needs your LinkedIn password — do that in LinkedCamp), CRM integrations, billing, team management and API-key changes. The agent will never ask for your LinkedIn password, cookies or verification codes.
Support
- Website: https://www.linkedcamp.com/
- Help: https://www.linkedcamp.com/contact
- Privacy: https://www.linkedcamp.com/privacy
- Terms: https://www.linkedcamp.com/terms
License
Proprietary. © LinkedCamp.
