AI investigate
loguro investigate runs an AI investigation against a single log ID. The result is cached, so re-running is free; bypass the cache with --no-cache.
loguro investigate 01KQ0E1ZTAP7YWFBK1S2K6XM2W
loguro investigate 01KQ0E1ZTAP7YWFBK1S2K6XM2W --no-cache
loguro investigate 01KQ0E1ZTAP7YWFBK1S2K6XM2W --raw # streamed text for pipesInvestigate alongside the log
loguro get --investigate prints the log with full context, then runs the investigation right after — handy when you want both in one scrollback:
loguro get 01KQ0E1ZTAP7YWFBK1S2K6XM2W --investigateFlags
| Flag | Description |
|---|---|
--no-cache | Force a fresh investigation, ignoring any cached result |
--raw | Output raw streamed text (no formatting), suitable for piping |
Quota
Each investigation consumes from your plan’s AI quota. Cached results don’t count — only new investigations and --no-cache re-runs do. Check your remaining quota with loguro usage.
Code-aware mode
When a GitHub integration is connected for the project, the AI investigator can search your repo to ground its explanation in real code. Connect it once with loguro integrations connect github (see the Integrations doc for full options).
Without GitHub connected, investigation works fine — it just relies on the log message + context only.