Store
Every LinkedIn agent, skill and workflow we ship.
Install one into Claude, Cursor or VS Code, or run it here on a hosted seat. Each listing says what it can do, which tool does it, and what it risks — filter on any of that.
Not sure which one? Answer a few questions and get a recommendation →
Filters · 1 on
Outcome
Type
Where it runs
Taken from what each product declares it ships for. Where a product declares nothing, it falls back to a rule: a hosted seat runs here, everything else is files and config any MCP client can load.
Risk
Status
Price
Bands are the published ladder. Priced-on-request products are excluded from a band — we do not know that they fit under it.
No products
Nothing matches that combination.
Every filter below is on at once. Take one off and the catalogue widens; the counts in the rail show what each one would return.
How to read the badges
Every listing states how it reaches LinkedIn and what that exposes you to. These are the only definitions; a product page does not get to soften them. Nothing here is sanctioned by LinkedIn, and no tool — ours or anyone else’s — can promise an account will not be restricted.
Risk level
- Low risk
- Reads only, at human pace. No writes to LinkedIn.
- Medium risk
- Acts through your own session or writes with a human approving each action. Restriction is unlikely but not impossible.
- High risk
- Automates actions LinkedIn watches for. Use a dedicated account and expect that it can be restricted.
Automation method
- Official API
- Goes through a documented API with a token you granted.
- Your LinkedIn session
- Acts as you, from your connected account, under pacing limits.
- Human approves
- Drafts the action; nothing irreversible happens until a person presses send.
- Read only
- Fetches and reasons. It cannot write anything to LinkedIn.