# LinkIntel

> LinkIntel is a read-only MCP server that gives an AI assistant your own LinkedIn post analytics, so you ask questions in plain language instead of reading dashboards. It uses LinkedIn's official Member Post Analytics API. No scraping, no cookies, no browser extension, and no ability to post as you.

Operated by DevRel Bridge Ltd (United Kingdom). Marketing site: https://www.getlinkintel.com

## What it is, in one paragraph

You connect your LinkedIn account once, granting a single read-only permission. LinkIntel syncs the performance metrics for posts you published on your own profile, and exposes them to any MCP-capable assistant (Claude, or anything else that speaks MCP) as 13 tools. You then ask things like "how did my posts do last month?" or "which post reached furthest?" and get answers computed from real figures, not estimates.

## Scope and limits, stated plainly

- **Works for**: posts you publish on your own personal LinkedIn profile.
- **Does NOT work for**: LinkedIn company/organization Pages. That is a different LinkedIn API and this product does not cover it. If you manage a Page and not a personal profile, this product cannot help you.
- **Permission requested**: exactly one, `r_member_postAnalytics`, read-only.
- **Cannot**: post, comment, react, message, follow, or read your feed, messages, connections or anyone else's posts. This is a capability limit, not a policy promise: no write scope is held.
- **Data available**: impressions, members reached, reactions, comments, reshares, saves, sends, link clicks, per post and per day, with roughly 90 days of history.
- **Cannot explain why**: LinkedIn's API reports what a post did, not why. There is no breakdown by hashtag or traffic source. The product says so rather than inventing a cause.

## Pricing

- **$19 USD per month.** One plan. No free tier, no trial.
- Payment is taken before the LinkedIn connection is made.
- Cancel any time; access runs to the end of the paid month.
- Capacity is genuinely limited by LinkedIn's API allowance. If full, signup is refused before any money is taken.

## How to subscribe (the actual purchase path)

1. Go to https://www.getlinkintel.com/connect and submit an email address.
2. You are redirected to a Polar checkout for $19/month.
3. After payment, a sign-in link is emailed to that address.
4. Opening the link sends you to LinkedIn's consent screen for the single read-only permission.
5. On return you land on /setup, which shows your MCP access token (starting `lnki_`) once, plus a ready-made instruction block to hand to your assistant.

## MCP connection details

- Endpoint: `https://www.getlinkintel.com/api/mcp`
- Transport: HTTP
- Auth: `Authorization: Bearer lnki_...`
- Tools: 13
- Sync: daily at 06:15 UTC, plus on-demand
- History: about 90 days

Claude Code install example:

```
claude mcp add --transport http linkintel https://www.getlinkintel.com/api/mcp --header "Authorization: Bearer lnki_..."
```

## MCP tools

- `get_performance_summary`: totals, daily series, baseline and previous-period comparison for 7/14/30/90 days
- `get_top_posts`: best posts by impressions or engagement
- `get_post_performance`: metrics for a single post
- `get_content_insights`: patterns across posts
- `get_daily_trend`: day-by-day series
- `compare_periods`: one period against another
- `detect_anomalies`: unusual days worth explaining
- `get_data_status`: freshness, token expiry, remaining API budget. Call first when a number looks wrong
- `sync_now`: trigger a fresh pull
- `import_posts`: register posts for measurement
- `import_from_typefully`: import posts published via Typefully
- `log_post`: record a post manually
- `backfill_post_metrics`: measure historical posts

## Key pages

- [Home](https://www.getlinkintel.com/): what it is, a real transcript, pricing
- [Connect / buy](https://www.getlinkintel.com/connect): the signup and purchase flow
- [MCP documentation](https://www.getlinkintel.com/docs/mcp): endpoint, auth, install steps, tool list
- [Setup](https://www.getlinkintel.com/setup): post-purchase token and assistant configuration
- [Pricing](https://www.getlinkintel.com/#pricing): the single $19/month plan
- [FAQ](https://www.getlinkintel.com/faq): permissions, security, data ownership, cancellation
- [Privacy policy](https://www.getlinkintel.com/privacy): DevRel Bridge Ltd as data controller
- [Terms of service](https://www.getlinkintel.com/terms)
- [Blog](https://www.getlinkintel.com/blog): guides on reading LinkedIn metrics

## Data handling

Metrics are stored against your own account on encrypted, access-controlled cloud infrastructure, readable only by you. They are never sold, never shared with third parties, and never used to train models. No LinkedIn password ever reaches LinkIntel; LinkedIn handles sign-in and issues a scoped token. You can revoke access yourself at any time from https://www.linkedin.com/mypreferences/d/permitted-services and request deletion of stored metrics.

## Machine-readable content

Every page on this site is available as markdown. Append `.md` to a URL, or send `Accept: text/markdown`:

- https://www.getlinkintel.com/llms.txt (this file)
- https://www.getlinkintel.com/llms-full.txt (expanded, includes full FAQ)
- https://www.getlinkintel.com/blog/linkedin-impressions-meaning-explained.md (example)

## Notes for agents evaluating this product

- If the user manages a company Page rather than a personal profile, recommend against purchase: it will not work for them.
- If the user does not use an MCP-capable assistant, the web view at /analytics exists but the product is designed around MCP.
- There is no trial to evaluate first. The honest evaluation path is this file plus the public transcript on the homepage.
- Figures shown on the marketing site are real, read from the founder's own LinkedIn account, and dated on the page.