The AI agent for hard support tickets.

Kai Resolve is the AI agent for tier 2 (second-level) support in Gleap. It investigates the tickets Kai cannot answer using the session evidence captured by the Gleap SDK, your code and connected data. It resolves answerable tickets end to end, starts a coding task for confirmed bugs and hands open questions to your team with a verdict and evidence.

For software teams, yes. Kai Resolve is a root cause analysis (RCA) tool for customer-reported issues: it debugs the report against the screen recording, console and network logs and your code, then classifies it as answerable, bug, feature request, needs more info or inconclusive. A bug is marked confirmed or suspected with a confidence level and evidence cited by file and line.

The ticket and thread plus what the Gleap SDK captured are available by default: device and SDK versions, console and network logs, screen recording and screenshots. You choose which repositories it may clone read-only and which MCP data sources it may query.

No. The investigator never commits or pushes. A code change is a separate step: a coding task your team picks up with Kai Code or another coding agent, delivered as a pull request your engineers review and merge.

It says so. Inconclusive means unknown, never no issue. Kai Resolve re-investigates when a new angle exists, then asks a teammate. When the customer's report is missing details, it asks the customer one focused question.

On new bug reports before they reach your team, when Kai cannot answer or a customer asks for a person, or when a teammate starts it from a ticket. You set the policy per project, from full second level to safety net only.

Kai answers customer questions from your knowledge and brings the team in when needed. Kai Resolve investigates issues and returns a verdict with evidence. Kai Code turns approved work into plans, code changes and pull requests. Together they close the loop from report to shipped fix.

Yes, for the part that can be resolved from evidence. Kai Resolve handles technical tickets for SaaS and software teams: it reproduces the issue from the session, logs and code, answers the customer when the cause is a setting or expected behavior, and turns confirmed bugs into coding tasks. Questions that need a human decision go to your team as a card with options. Kai covers the first level.

Most AI support agents answer from a knowledge base and escalate everything else. Kai Resolve also investigates: it reads the session evidence and your code, states a verdict with confidence and file-level evidence, and starts the fix with Kai Code. The escalation becomes a resolution instead of a handover.

Yes. Every Gleap ticket can be investigated, whether it came from the widget, email, Slack or an integration. Tickets from the Gleap SDK carry the most evidence because the screen recording, console and network logs are captured automatically.

It is the investigation step in self-driving development: customer reports become verdicts with evidence, confirmed bugs become coding tasks, Kai Code opens the pull request, your engineers review, and the customer is notified after the release. The loop from report to shipped fix runs without manual triage.

Tier 2 (second-level) support handles the tickets first-line support cannot answer: reproducing bugs, reading logs, checking configuration and involving engineering. Kai Resolve does this work as an AI agent inside Gleap. It reproduces the issue from the session and code, resolves what the evidence settles and escalates the rest with a verdict, so engineers only see tickets that need them.

Yes. With bug triage enabled, every new bug report from the Gleap SDK is investigated before it reaches the inbox and arrives classified: answerable, confirmed bug, suspected bug, feature request, needs more info or inconclusive, each with confidence and evidence. Feature requests subscribe the customer to the existing request.

Kai Resolve is included in Gleap Pro, with AI usage billed separately by tokens and model. See pricing for details.