GlinkBot documentation

Approvals and inspector

Stay in control when agents ask questions, request access, or perform risky work.

GlinkBot renders structured agent events as native conversation cards. Questions, choices, permission requests, tool activity, and failures remain attached to the task that produced them.

Approval cards

When an engine supports approvals, GlinkBot surfaces the request instead of hiding it in terminal output. Read the requested action and scope before approving it. Denying an action sends control back to the agent so it can choose a safer path.

Access and outbound actions

GlinkBot defaults to Ask. Review permissions before increasing access. Engine approvals and GlinkBot's independent outbound consent are separate: Full access does not bypass the outbound grant or daily limit. Sending external messages requires the applicable consent and remaining allowance.

Review any proposed changes to bots, skills, working folders, models or routines. Where a change offers Undo, it may no longer be reversible after related state changes. Deleting a bot has no Undo.

Inspector

Use Inspector when you need more detail than the conversation provides. It is useful for following tool calls, reconnect attempts, engine output, and long-running work.

When a run disconnects

GlinkBot retries transient streaming failures. If retries end in an authentication or configuration error, fix the underlying engine rather than repeatedly retrying the task. See Agent engine troubleshooting.

GlinkBot documentation, reviewed for the 0.0.4 codebase. Public download availability can differ. Licenses and acknowledgments.