Preview and debug

Use the Agent workspace to test an optional preset in a Preview Session.

Preview

Enter a representative task in the left pane. Test:

a normal request and expected output;
a request that uses each attached Skill or MCP server;
file input and generated artifacts;
missing information and clarification behavior;
tool, credential, or Environment failure paths.

Saved preset edits apply to new Sessions. Start a new Preview when testing changed configuration. Cloud debug Previews may be cleaned up after 30 days without message, Run or file activity; opening the console or reading history does not renew that period. Formal API Sessions do not use this expiry rule.

Logs

Open Logs to inspect session and execution history. Use it to correlate runtime startup, tool use, and failures. Secrets should not appear in logs; if a raw credential is present in user-authored content, remove it before sharing diagnostics.

Cost

Open Cost for Agent-level spend estimates, model mix, and recent usage events. Filter production and debug use separately. Cost values are estimates from recorded model calls, not Provider invoices.

Session maintenance

Maintenance actions target the selected Session. Restarting its Driver or recreating its sandbox preserves the committed workspace and native continuation; a different Session does not inherit that state. Start a new Session for unrelated work. A failed or incomplete checkpoint must be resolved before continuation or maintenance can use it.

Diagnose common failures

Runtime needs a key: configure the matching Provider in this Project.
Model is unavailable: verify the Provider's model list and endpoint compatibility.
MCP tool fails: confirm the connection is enabled and authorized, then retry a direct tool task.
Environment startup fails: check exact package versions, setup script, and required variables.
A saved edit is absent from an existing Session: configuration is fixed when a Session starts. Start a new Preview to test changed settings.