Описание
LinkedIn Agent skills from contentin.io
README
Agent skills and MCP configs that connect Claude, ChatGPT, Claude Code, Cursor, Gemini CLI, OpenClaw, Codex, n8n or any other MCP client to ContentIn, so the agent you already run writes LinkedIn posts in your voice from your own material, schedules and publishes them through LinkedIn's official API after you confirm, measures them, and (on Agent Pro) shows you who engaged.
Your agent can already write. On its own it writes like everyone else's agent. Connected to ContentIn it gets your voice and your own stories and numbers (so it stops inventing), what worked on your account, the people who engaged with your posts scored against your ideal customer, and a safe last mile: nothing goes out without your confirmation.
Launching on Product Hunt on 26 November 2026. Follow ContentIn on Product Hunt to get notified when it goes live.
What is in this repo
| Path | What |
|---|---|
skills/contentin-linkedin/SKILL.md |
The main skill: when to use each of the 13 tools, which plan each needs, the capture, voice, draft, confirm, schedule workflow, the confirmation rule, and what to tell the user when a tool refuses. |
skills/contentin-linkedin-leads/SKILL.md |
The lead workflow (Agent Pro or Pro): list the people who engaged, pick, draft a reply or note the user sends themselves. Never sends anything. |
.claude-plugin/ |
Plugin and marketplace manifests, so this repo installs as a Claude Code plugin that brings the MCP server and both skills. |
plugin.json, mcp.json |
The same plugin in the Agent Plugins format read by Cursor's plugin directory and OpenAI's plugin tooling. |
.mcp.json |
The MCP server in the shape Claude Code and Cursor read from a project root. |
gemini-extension.json |
The same package as a Gemini CLI extension: gemini extensions install https://github.com/Kinzi/contentin-agent-skills. |
mcp/ |
Ready config snippets per client, in OAuth form and in API-key form. |
No code. The skills are plain Markdown following the Agent Skills standard. The MCP server runs at https://mcp.contentin.io/mcp-server (Streamable HTTP).
Who it is for
People who run AI agents and automations and post on LinkedIn: founders, solo consultants, coaches, fractional executives, and the people who build LinkedIn workflows in n8n or Claude Code for them.
1. Get a ContentIn account
Create the account at app.contentin.io/signup?product=agent, or connect from your AI client and create it on the sign-in page that opens. Paste your LinkedIn profile URL; ContentIn reads your published posts and shows you how you write. Account, connection and voice profile are free, no card.
The tools that read your posts, write, schedule and publish need a plan: Agent at $29 or Agent Pro at $49 a month per LinkedIn profile, each with a 7-day trial (card required). Already on ContentIn Growth or Pro? The same tools are included (Growth = Agent tools, Pro = Agent Pro tools), nothing to buy.
2. Connect the MCP server
Server URL for every client: https://mcp.contentin.io/mcp-server
| Client | How |
|---|---|
| Claude Code | claude mcp add --transport http contentin https://mcp.contentin.io/mcp-server then /mcp, select contentin, Authenticate. Or install the plugin (next section). |
| Claude.ai and Claude Desktop | Settings, Connectors, Add custom connector. Paste the URL, leave client id and secret empty, Connect, sign in with ContentIn. |
| ChatGPT | Settings, Apps, turn on developer mode, create a new app or connector, enter the URL, choose OAuth. |
| Cursor | Copy mcp/cursor/mcp.oauth.json to .cursor/mcp.json (project) or ~/.cursor/mcp.json, then sign in from the contentin entry in Cursor's MCP settings. |
| VS Code | Copy mcp/vscode/mcp.oauth.json to .vscode/mcp.json, or run "MCP: Add Server" and paste the URL. |
| Gemini CLI | gemini extensions install https://github.com/Kinzi/contentin-agent-skills (brings the server and both skills), then /mcp auth contentin. Or merge mcp/gemini-cli/settings.oauth.json into ~/.gemini/settings.json. |
| OpenClaw | openclaw mcp set contentin '{"url":"https://mcp.contentin.io/mcp-server","transport":"streamable-http","auth":"oauth"}' then openclaw mcp login contentin. Or merge the JSON5 fragment in mcp/openclaw/openclaw.oauth.md. |
| Codex CLI | Add the TOML block from mcp/codex/config.oauth.md to ~/.codex/config.toml, then codex mcp login contentin. |
| n8n and scripts | Create an API key in ContentIn under Profiles, Connections ("Connect AI clients"). In n8n add an MCP Client node, endpoint = the URL, transport HTTP Streamable, Authentication = Header Auth with name X-MCP-Key and the key as value. |
API keys work in every client that accepts headers: send X-MCP-Key: <your key>. The *.apikey.* files in mcp/ show the exact shape per client. Each key or OAuth grant is bound to one ContentIn profile, which you pick when you connect. Revoke any time under Profiles, Connections, Connected apps.
Full setup guide: https://contentin.io/linkedin-mcp-server/setup/
3. Install the skills
With the skills CLI (Claude Code, Cursor, Codex, Gemini CLI, OpenClaw, Copilot, Windsurf and others; needs Node.js 22.20+):
npx skills add Kinzi/contentin-agent-skills
Pick one skill only:
npx skills add Kinzi/contentin-agent-skills --skill contentin-linkedin
As a Claude Code plugin (brings the MCP server too):
claude plugin marketplace add Kinzi/contentin-agent-skills
claude plugin install contentin-linkedin@contentin
Then run /mcp once to sign in. Other agents: copy the skills/<name>/ folder into the place your agent reads skills from (for example .claude/skills/, .cursor/skills/, .agents/skills/, or skills/ in an OpenClaw workspace).
The 13 tools
| Tool | Plan |
|---|---|
get_account_status: setup, plan, LinkedIn connection and what each tool is waiting for |
any connected account |
get_voice_context: your voice, rules, saved material, best posts and strategy, served before the agent writes |
Agent |
list_posts, get_post_analytics: your drafts, queue, published posts and per-post results |
Agent |
capture_substance: save a story, opinion or number to your material bank |
Agent |
write_post_in_my_voice, repurpose_post, generate_ideas: ContentIn's own writer (Agent: 30 posts and 50 ideas a month; Agent Pro: 100 and 200) |
Agent |
attach_images: up to five images on a draft |
Agent |
schedule_post, publish_post: show the exact text, wait for your yes, then go through LinkedIn's official API |
Agent, LinkedIn connected |
list_leads, list_my_comments: people who engaged with your posts with ICP fit score; comments you left elsewhere |
Agent Pro or Pro |
Agent plans allow at most 3 scheduled or published posts per profile per day.
Safety
- Scheduling and publishing are two-step. The tool returns the exact text and a short-lived single-use token; nothing goes out until you confirm in your own words.
- One grant = one profile. The agent cannot reach other profiles, billing or LinkedIn credentials.
- The leads skill drafts only. It never sends comments, messages or connection requests.
- See
SECURITY.mdfor reporting and for what is logged.
Links
- Hub: https://contentin.io/linkedin-for-ai-agents/
- Setup guide: https://contentin.io/linkedin-mcp-server/setup/
- Privacy: https://www.iubenda.com/privacy-policy/58097432
- Terms: https://contentin.io/blog/linkedin-mcp-terms-of-service/
- Support: [email protected]
Licence
MIT, ContentIn LLC. See LICENSE. The skills and configs in this repo are free to copy and adapt; the ContentIn service itself is a paid subscription governed by its own terms.
Установка Contentin Agent Skills
У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.
▸ github.com/Kinzi/contentin-agent-skillsFAQ
Contentin Agent Skills MCP бесплатный?
Да, Contentin Agent Skills MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Contentin Agent Skills?
Нет, Contentin Agent Skills работает без API-ключей и переменных окружения.
Contentin Agent Skills — hosted или self-hosted?
Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.
Как установить Contentin Agent Skills в Claude Desktop, Claude Code или Cursor?
Открой Contentin Agent Skills на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Opencode Omniroute Plugin
OpenCode plugin for the OmniRoute AI Gateway. Drives dynamic model discovery, /connect auth flow, and multi-instance OmniRoute providers via the official @openc
автор: GitHub ActionsRoblox Studio
Enables AI coding tools to control Roblox Studio for workspace exploration, instance manipulation, and script management. It provides tools for playtesting, sce
автор: paralovAntigravity Drawio
MCP server that enables AI assistants to create, parse, render, and validate Draw.io diagrams programmatically.
автор: S-SUJAN-SFetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzMCP-Agent
A simple, composable framework to build agents using Model Context Protocol by [LastMile AI](https://www.lastmileai.dev)
автор: lastmile-aiSpring AI MCP Client
Provides auto-configuration for MCP client functionality in Spring Boot applications.
mcp.natoma.ai
A Hosted MCP Platform to discover, install, manage and deploy MCP servers by [Natoma Labs](https://www.natoma.ai)
Compare Contentin Agent Skills with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
