Is Crispy safe to use?

Crispy is designed so an AI agent works your LinkedIn account at a human pace, not a bot pace. Several protection layers are always on.

• Daily caps per action category. Every write action (invitations, messages, posts, engagement, and more) counts against a per-account daily cap and is blocked once the cap is reached. Counters reset at midnight UTC.
• Invitation pacing. Your daily invitation budget is spread across 8 hours rather than being spendable all at once, so a day's invitations cannot go out in a few minutes.
• Warm-up for new accounts. An opt-in ramp that starts a new account at low limits and raises them gradually over about two weeks.
• Business hours. Campaign sends only fire inside a send window, by default 9:00 to 18:00 in the account's timezone.
• Natural variation. You can set a minimum and maximum per category, and Crispy picks a different daily cap inside that range each day, so output does not hit the same number every day.
• Block list. Anyone on your do-not-contact list is skipped automatically by campaigns and flagged in search results.
• Permission scopes. You can restrict what an agent or API key is allowed to do, down to read-only.

Most read tools, like listing conversations or pulling analytics, have no safety limit at all. Profile views and searches are capped because they generate real LinkedIn activity.

What you can do on top:

• Run warm-up on any genuinely new account before launching campaigns.
• Keep daily activity comfortably below the caps rather than at them.
• Never run a second automation tool on the same LinkedIn account at the same time. Two tools cannot see each other's activity, so combined volume can spike past comfortable levels.

No tool can make LinkedIn automation zero-risk. Crispy's job is to keep your account's activity paced and shaped like a person's, and every layer above exists for that reason.