Grok Bot inference hooks let your security server block a prompt before the model sees it
3 weeks ago
Grok Bot inference hooks send each model request, signed, to your organization's AI security server, which allows or denies it before the model runs. A beta for every Enterprise team: Shadow mode, fail-open or fail-closed modes, what blocked members see, and what the hook doesn't cover. Plus Pages, and Apply to MCP servers for the network allowlist. https://docs.x.ai/grok-bot/inference-hooks https://docs.x.ai/grok-bot/security https://docs.x.ai/grok-bot/files-and-results https://docs.x.ai/grok-bot/teams-and-enterprises https://docs.x.ai/developers/release-notes.md
Ask about this presentation
Chapters
Show transcriptHide transcript
Grok Bot can now send each model request to your organization's AI security server. The request is signed, and your server allows or denies it before the model runs. It's a beta for every Enterprise team.
The check covers what the Bot picks up, as well as what members type. Web pages, app results and command output reach your server before the next model call, so your data loss policy can stop them.
A new configuration starts in Shadow mode. Your server sees every covered request and denials go to the audit log, while members keep working. Their conversations still reach your server, so tell members first.
Once the verdicts look right, you pick a mode. Allow the Request lets traffic through when your server times out or errors, and Block the Request stops it. A failure never counts as a deny.
On a deny, the sender sees Blocked by your organization's policy, with your server's reason. Edit puts the text back so they can send something different, and Reset clears the conversation.
A tool call reaches your server only after it has run, and that includes the Bot's replies to members. So a deny stops the next model call. Keep the network policy and Auto Review in place to control actions.
Coverage stops at Grok Bot. Cloud Agents, editor agents, the CLI and the Agent SDK go to the model unchecked. And with a hook on, Grok Bot skips some early model calls, so replies can start more slowly.
A Page is a styled report, dashboard or tracker the Bot builds, like signups for the last 30 days. Self-serve plans can share one publicly, Enterprise plans with chosen people and teams, and routines can keep it updated.
Grok Bot reaches connected apps from Cursor's servers, so a network allowlist alone misses them. Turn on Apply to MCP servers and every app, GitHub and X included, must match an allowed host. It's off by default.
Setup steps, the request format and the audit events are in the Grok Bot inference hooks guide. Which security server would your team point Grok Bot at?
Up next
1:52Grok Bot's Vault lets Bots sign in and pay without seeing your passwords
@grok-bot
1 day ago
1:06Grok Bot now acts only as you, and keeps your passwords out of the model
@grok-bot
1 day ago
2:29Grok Bot now signs in with Okta FastPass on its own computers
@grok-bot
2 days ago
2:19Grok Bot now shares one Bot with your whole team, in Slack and Teams
@grok-bot
2 days ago
1:59Grok Bot Team Bots now learn for the team and join Microsoft Teams
@grok-bot
3 days ago
1:56Grok Bot buys your conference pass from memory and asks before it emails finance
@grok-bot
1 week ago
1:04Grok Bot now shuts idle computers after 30 days without a message
@grok-bot
1 week ago
1:58Grok Bot admins can now wipe a member's Bots and data in one step
@grok-bot
3 weeks ago
1:47Grok Bot lets Enterprise team admins recreate or terminate Bot computers in bulk
@grok-bot
1 month ago
2:52Grok Bot now deploys to your whole organization with Jamf, Intune and Ansible
@grok-bot
1 month ago
1:19Grok Bot plugins now install from Connect apps
@grok-bot
1 month ago
1:43Grok Bot Team Bots get managers, and Bots can read X
@grok-bot
1 month ago










