RXed AI News

Your AI intelligence briefing
2.7/10 · Agent-hostile Assessed 2026-07-12 · AFI rubric v0.1

LinkedIn — how agent-ready is it?

The most agent-hostile mainstream platform: five partner-gated API tiers, months-long manual approvals, litigated anti-scraping enforcement, and no agent story at all.

API surface — 3/10

Five separate gated tiers; core profile and network data are inaccessible outside narrow partner programs; Sales Navigator API closed to new partners.

MCP / agent protocol — 1/10

No official MCP server and no sanctioned agent integration path.

Docs quality — 5/10

Microsoft Learn documentation is competent — for APIs most developers will never be approved to use.

Auth friction — 2/10

Marketing Developer Platform approval is manual with no published timeline — realistically 4 weeks to 4 months, frequently rejected.

Rate limits & pricing fairness — 3/10

Tier-dependent and tight; rejected use cases explicitly include job-search automation and third-party profile analytics.

ToS stance on agents — 1.5/10

Scraping prohibited and litigated; automation against the UI banned; agent access is effectively a ToS violation by default.

Machine-readable output — 2/10

No public structured data; aggressive anti-bot walls; content locked behind login.

Stability — 4/10

The closed tiers are at least consistently closed; partner programs shrink rather than grow (SNAP closed to new applicants).

Sources

Point-in-time assessment — platforms change their API terms often (that volatility is itself scored under Stability).