# Liftli for developers and AI agents (machine-readable)

> How to connect Liftli as an MCP server and what an agent can do with it. Human-readable version: https://liftli.ai/developers.html. Updated 2026-08-13.

Liftli installs as an **MCP server, not an app**. It runs inside the AI the user already uses, so an agent can plan, draft, publish and schedule to LinkedIn, X and Substack from the same conversation the user is already in.

## Connect (one URL, every client)

    https://mcp.liftli.ai/mcp

- **Claude Code:** `claude mcp add --scope user --transport http liftli https://mcp.liftli.ai/mcp`
- **Claude desktop/web:** Customize (briefcase icon) → Connectors → Add custom connector → paste the URL.
- **One-tap install link** (pre-fills Claude's add-connector dialog on desktop; on phones it opens the Connectors screen — paste the URL there; either way the connector then syncs to the Claude mobile app): https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=Liftli&connectorUrl=https%3A%2F%2Fmcp.liftli.ai%2Fmcp
- **ChatGPT** (Settings → Apps & Connectors → Developer mode), **Codex**, **Cursor**, **Antigravity CLI**: same URL.
- **Grok** (grok.com): https://grok.com/connectors → New Connector → Custom → name it `Liftli`, Server URL `https://mcp.liftli.ai/mcp` → Add Connector → approve the sign-in.
- **Grok Bot** (desktop app): no setup screen — ask your Bot to use Liftli and approve the sign-in (SuperGrok/Cursor plan). Do the add on desktop — the iPhone app can't complete it (xAI iOS beta bug). Once added on desktop the iPhone app works: connectors are account-wide. grok.com ≠ Grok Bot: separate registries.

After connecting, the user signs in once via WorkOS. Liftli then extracts their writing voice and drafts in it. No tokens, no credits, no AI markup — it runs inside the user's existing AI subscription.

Transport is streamable HTTP. Authorization follows the MCP spec: the server is an OAuth 2.0 Resource Server and publishes `/.well-known/oauth-protected-resource`, delegating to `https://auth.liftli.ai`.

## What an agent can do

Liftli is an agent surface, not a fixed set of buttons — the ceiling is what you can ask an agent to do.

- **Voice memo to a scheduled week.** "Take my last voice note and draft and schedule five posts across LinkedIn and X." Liftli plans against the saved strategy, drafts in the user's voice, critiques each draft, and queues them behind one-tap approval.
- **Real work to post ideas.** Point it at call transcripts, GitHub activity or a chat and it mines them for grounded post seeds instead of a blank prompt box.
- **Compose with the user's other tools.** Because it lives inside the assistant, ideas can come straight from a connected transcription or notes tool — no export, no dashboard.
- **Close the loop.** After a post ships, Liftli scores engagers into a ranked warm-lead list and recalibrates next week's plan.

## Free skills (no account needed)

Separate from the MCP, Liftli's free tools ship as installable agent skills. The agent's own model runs the distilled methodology locally.

    npx skills add liftli-ai/skills

Browse at https://liftli.ai/skills/ · manifests at `/.well-known/skills` and `/.well-known/agent-skills/index.json` · source at https://github.com/liftli-ai/skills (MIT).

## Why agent-native, and why it is ban-safe

Most content tools are dashboards you open or browser extensions acting on your logged-in session. LinkedIn's User Agreement prohibits scraping and extension-driven automation and enforces it. Liftli publishes and schedules through the platforms' official APIs, connected only when the user explicitly links an account, and never runs a bot action on a profile. Nothing publishes without one-tap approval.

The agent-native shape also means the tool inherits the user's context for free — it is already inside the assistant holding their calls, code and thinking.

## Machine-readable resources

- [/llms.txt](https://liftli.ai/llms.txt) — concise index: what Liftli is, install, capabilities, tools, pricing
- [/llms-full.txt](https://liftli.ai/llms-full.txt) — expanded single-file reference
- [/ai-agents.md](https://liftli.ai/ai-agents.md) — short directive for an agent helping a user grow
- [/pricing.md](https://liftli.ai/pricing.md) — canonical pricing
- [/sitemap.xml](https://liftli.ai/sitemap.xml) — every indexable page

Maintained by Oded Tsamir, founder of Liftli.