Thinking: gate xhigh by model
This commit is contained in:
committed by
Peter Steinberger
parent
f50e06a1b6
commit
a3641526ab
@@ -38,7 +38,7 @@ clawdbot agent --to +15555550123 --message "Summon reply" --deliver
|
||||
- `--local`: run locally (requires provider keys in your shell)
|
||||
- `--deliver`: send the reply to the chosen provider (requires `--to`)
|
||||
- `--provider`: `whatsapp|telegram|discord|slack|signal|imessage` (default: `whatsapp`)
|
||||
- `--thinking <off|minimal|low|medium|high>`: persist thinking level
|
||||
- `--thinking <off|minimal|low|medium|high|xhigh>`: persist thinking level (GPT-5.2 + Codex models only)
|
||||
- `--verbose <on|off>`: persist verbose level
|
||||
- `--timeout <seconds>`: override agent timeout
|
||||
- `--json`: output structured JSON
|
||||
|
||||
Reference in New Issue
Block a user