# What is the LinkedIn API?

Canonical page: https://depost.ai/glossary/linkedin-api

**The LinkedIn API:** The LinkedIn API is the set of interfaces LinkedIn publishes for approved applications to act on a member’s behalf: the member grants specific permissions on a LinkedIn authorization screen, the application never receives a password, and access can be revoked at any time.

Also called: LinkedIn official API, LinkedIn developer API, LinkedIn integration.

The mechanics are worth knowing, because they are what the safety question turns on. An application registers with LinkedIn and is reviewed. When you connect it, LinkedIn shows you what it is asking for and issues the application a token if you agree. Your password never reaches it, and the connection is listed in your LinkedIn settings, where you can end it.

Access is deliberately narrower than the website. Publishing to your own account and reading how your own posts performed are the kinds of things approved applications can do. Bulk access to other members’ data is not on offer, which is why tools that sell contact-scraping cannot be doing it this way.

Some absences surprise people. LinkedIn provides no way for an application to post or pin a comment on your behalf, so a first comment has to be pasted by a person at publication. A tool claiming to schedule or pin one is describing a capability the platform does not expose.

The question this actually answers is not technical. It is whether the software is a participant in LinkedIn’s ecosystem, known to it and accountable, or an imitation of you operating from inside your signed-in browser session. Only one of those puts your account at risk.

Revocation is the practical test of the difference. If you can end a tool’s access from LinkedIn’s own settings, it was connected properly. If ending it means changing your password, it was never a connection in the first place.

## In Depost AI

Connecting Depost AI is this flow: LinkedIn asks whether you agree, and it works from the token you granted. It reads the history and analytics of that account and publishes the posts you approved, and you can revoke it from LinkedIn at any time without touching your password.

## Questions people ask

### Does connecting a tool through the LinkedIn API give it my password?

No. You authorize it on a LinkedIn screen, and the application receives a token rather than credentials. You can revoke that token from your LinkedIn settings, which is something a password-based tool cannot offer.

### Can any tool use the LinkedIn API?

No. Access is granted to applications LinkedIn has approved, and the available capabilities are limited. That approval step is precisely why some tools drive a signed-in browser session instead.

## Read next

- [Scheduling and publishing](https://depost.ai/features/scheduling)
- [Analytics that drive what you post](https://depost.ai/features/analytics)

## Related terms

- [LinkedIn automation](https://depost.ai/glossary/linkedin-automation): LinkedIn automation is software carrying out actions on a member’s behalf, and it covers two mechanisms that are routinely confused: applications authorized through LinkedIn’s official API, which act when a person tells them to, and bots that sign in as the member and click for them.
- [Account restriction](https://depost.ai/glossary/account-restriction): An account restriction is a limit LinkedIn places on what an account can do, applied when its activity looks automated or otherwise inconsistent with the User Agreement, and ranging from a single feature refusing to work to permanent closure.
- [Model Context Protocol (MCP)](https://depost.ai/glossary/model-context-protocol): The Model Context Protocol is an open standard that lets an AI assistant connect to external tools and data sources, so the assistant can read from and act in another system without that system building a custom integration for each assistant.

---

Cite as Depost AI (https://depost.ai/glossary/linkedin-api). Index of every page: https://depost.ai/llms.txt. Full reference: https://depost.ai/llms-full.txt.
