What causes "chatgpt detail" in jackwener/OpenCLI?

Re-run with a higher --timeout value, as the error hint suggests

jackwener/OpenCLI · error · TimeoutError

chatgpt detail

Error message

chatgpt detail

What it means

The chatgpt conversation-detail/wait routine polls the page for the conversation to finish and reach a stable state (unchanged lastKey for the stability window). If it does not stabilize within timeoutSeconds, a TimeoutError is thrown with a hint to raise --timeout.

Source

Thrown at clis/chatgpt/utils.js:1305

                            wantMarkdown,
                            generating: false,
                            stableSeconds: elapsedSeconds,
                        }),
                        generating: false,
                    };
                }
            } else {
                lastKey = key;
                stableStartedAt = Date.now();
            }
        } else {
            lastKey = key;
            stableStartedAt = 0;
        }
        await page.sleep(3);
    }

    throw new TimeoutError(
        'chatgpt detail',
        timeoutSeconds,
        'Conversation did not finish or stabilize before timeout. Re-run with a higher --timeout if it is still generating.',
    );
}

function normalizeDeepResearchText(value) {
    return String(value || '')
        .replace(/\u00a0/g, ' ')
        .replace(/[ \t]+\n/g, '\n')
        .replace(/\n{3,}/g, '\n\n')
        .trim();
}

function looksLikeDeepResearchReport(text) {
    const normalized = normalizeDeepResearchText(text);
    if (normalized.length < 500) return false;
    return /(^|\n)\s*#{1,3}\s+\S|Sources|References|参考|来源|结论|建议|Executive Summary|摘要/i.test(normalized);

View on GitHub (pinned to 49907e53dc)

Solutions

Re-run with a higher --timeout value, as the error hint suggests
Check the ChatGPT session is still logged in and not blocked by captcha/rate limits
Retry on a more stable network connection
If the conversation is genuinely stuck, send a new message to unstick it before re-running

Example fix

// before
opencli chatgpt read --url 'https://chatgpt.com/c/abc' --timeout 30
// after
opencli chatgpt read --url 'https://chatgpt.com/c/abc' --timeout 120

Defensive patterns

Strategy: retry

Validate before calling

if (typeof timeoutSeconds !== 'number' || timeoutSeconds < 60) {
  timeoutSeconds = 120; // long generations need headroom
}

Try / catch

try {
  const detail = await getChatGPTDetail(page, { timeoutSeconds: 120 });
} catch (err) {
  if (err instanceof TimeoutError && err.operation === 'chatgpt detail') {
    const detail = await getChatGPTDetail(page, { timeoutSeconds: 300 });
  } else throw err;
}

Prevention

Set --timeout generously (120s+) for long conversations
Check the session is authenticated and not rate-limited before starting long waits
Retry once with a doubled timeout before giving up
Verify the conversation has actually stopped streaming if it times out repeatedly

When it happens

Trigger: Calling a chatgpt command that waits for a completed/stable conversation while the model is still generating a long response, the stream stalls, the session is rate-limited or logged out mid-run, or the stability detection key keeps changing due to live UI updates.

Common situations: Very long generations exceeding the default --timeout; slow network or throttled account; ChatGPT UI showing a captcha/login prompt so the conversation never progresses; polling a conversation that keeps auto-updating.

Related errors

AI-assisted analysis of jackwener/OpenCLI@49907e53dc (2026-08-29). Data as JSON: /api/errors/b2daee1c4ce436fe. Report an issue: GitHub.