jackwener/OpenCLI · error · CommandExecutionError

Failed to send Qianwen prompt

Error message

Failed to send Qianwen prompt

What it means

clis/qwen/send.js calls sendMessage(page, prompt) after applying think/research toggles; if send?.ok is falsy it checks hasLoginGate (throwing an auth error if present), otherwise throws CommandExecutionError(send?.reason || 'Failed to send Qianwen prompt'). The text prompt was not delivered to the chat.

Source

Thrown at clis/qwen/send.js:51

        const prompt = String(kwargs.prompt || '').trim();
        if (!prompt) throw new ArgumentError('prompt is required');
        const startFresh = normalizeBooleanFlag(kwargs.new, false);
        const useThink = normalizeBooleanFlag(kwargs.think, false);
        const useResearch = normalizeBooleanFlag(kwargs.research, false);

        await ensureOnQianwen(page);
        await dismissLoginModal(page);
        if (startFresh) {
            await startNewChat(page);
            await dismissLoginModal(page);
        }
        if (useThink) await setFeatureToggle(page, 'think', true);
        if (useResearch) await setFeatureToggle(page, 'research', true);

        const send = await sendMessage(page, prompt);
        if (!send?.ok) {
            if (await hasLoginGate(page)) throw authRequired();
            throw new CommandExecutionError(send?.reason || 'Failed to send Qianwen prompt');
        }
        return [{ Status: 'sent', Prompt: prompt }];
    },
});

View on GitHub (pinned to 49907e53dc)

Solutions

1. Confirm login state in the automation browser and retry (a visible login gate means re-authenticate)
2. Retry after a slightly longer initial wait so the composer is fully rendered
3. Disable --think/--research toggles to isolate whether a toggle breaks the send
4. If persistent, the sendMessage selectors likely need updating after a Qianwen UI change

Example fix

// before
clis qwen send --prompt 'hi' --research true   # fails: feature/composer unavailable
// after
clis qwen send --prompt 'hi'   # send without toggles, add toggles back incrementally

null

null

try {
  await runQianwenSend({ prompt, think, research });
} catch (e) {
  if (/Failed to send Qianwen prompt/.test(e.message)) {
    if (think || research) {
      return runQianwenSend({ prompt });   // retry without feature toggles
    }
    await ensureLoggedIn(page);
    await sleep(2000);
    return runQianwenSend({ prompt });     // retry once
  } else throw e;
}

When it happens

Trigger: The composer input or send button was not found (DOM change, page still loading); the session is logged out but the gate check misses it; DeepThink/DeepResearch toggle interaction left the composer disabled; send timed out silently giving no reason.

Common situations: Slow page load so sendMessage runs before the composer renders; expired login cookies; site update renaming the send button selector; toggling research mode on an account where the feature is unavailable.

Related errors

• Failed to send Qianwen image prompt
• Failed to extract Booking.com cards: ${err?.message || err}
• Failed to send message
• coupang add-to-cart evaluation failed: ${error?.message || e
• coupang search

AI-assisted analysis of jackwener/OpenCLI@49907e53dc (2026-08-29). Data as JSON: /api/errors/e65084d93cb49caf. Report an issue: GitHub.