#gemini 3 flash from google ai studio api key isnt working as an agent
1 messages · Page 1 of 1 (latest)
What youâre describing is usually (a) âreasoning is being shownâ and/or (b) the model is planning but not actually tool-calling.
1) Stop it from âexplaining what itâs thinkingâ
In the channel/thread where you talk to the agent, send these as standalone messages:
/reasoning offâ hides the âReasoning:â / thinking blocks/verbose offâ stops the extra tool/debug bubbles (if you enabled them)
Docs: https://docs.openclaw.ai/tools/thinking (covers /reasoning, /verbose, /think)
If itâs still âthinking out loudâ inside the normal answer text, thatâs not OpenClawâs reasoning display â thatâs the Gemini model outputting it directly. In that case the fix is prompt/model-side (e.g. add âDonât reveal internal reasoning, just give the resultâ to your agent/system prompt, or switch to a model that supports separated/hidden reasoning).
2) âI need to keep prompting it to do thingsâ
That usually happens when tools are unavailable/blocked or the model is weak at tool-calling.
Quick checks:
- Do you see errors like âTool ⊠not allowedâ or âTool ⊠not foundâ in your gateway logs?
- When you ask it to do something concrete (web fetch, browser, exec, etc.), does it actually call tools or just describe steps?
If you paste:
openclaw --version- what exact model/provider youâre using (the
/model statusoutput is perfect) - a short snippet of a reply where it âthinksâ + the relevant gateway log lines
âŠI can tell you which of the above cases youâre in and what to change.