jackwener/OpenCLI · error · CommandExecutionError

ChatGPT conversation payload id mismatch: expected ${expecte

Error message

ChatGPT conversation payload id mismatch: expected ${expectedConversationId}, got ${payloadConversationId}.

What it means

When the caller passes expectedConversationId, the payload's own id (from conversation_id, conversationId, or id) must match it exactly. A mismatch means the response body belongs to a different conversation than the one requested, which would leak or attach Deep Research progress to the wrong thread, so the library throws.

Source

Thrown at clis/chatgpt/utils.js:1535

            report,
            html: '',
            method: source,
            sources: extractDeepResearchSourcesFromReportMessage(reportMessage),
            widgetStatus: String(widgetStateObject.status || ''),
            reportMessageId: String(reportMessage?.id || ''),
            reportLength: report.length,
        };
    }
    return buildDeepResearchProgressResult(widgetStateObject, pickFirstObject(responseMetadata), source);
}

function extractDeepResearchFromConversationPayload(payload, { expectedConversationId = '' } = {}) {
    if (!payload || typeof payload !== 'object' || Array.isArray(payload)) {
        throw new CommandExecutionError('Malformed ChatGPT conversation payload for Deep Research extraction.');
    }
    const payloadConversationId = String(payload.conversation_id || payload.conversationId || payload.id || '').trim();
    if (expectedConversationId && payloadConversationId && payloadConversationId !== expectedConversationId) {
        throw new CommandExecutionError(
            `ChatGPT conversation payload id mismatch: expected ${expectedConversationId}, got ${payloadConversationId}.`,
        );
    }
    const mapping = payload?.mapping && typeof payload.mapping === 'object' ? payload.mapping : {};
    if (!payload.mapping || typeof payload.mapping !== 'object' || Array.isArray(payload.mapping)) {
        throw new CommandExecutionError('Malformed ChatGPT conversation payload for Deep Research extraction: missing mapping.');
    }
    const candidates = [];
    for (const [messageId, node] of Object.entries(mapping)) {
        const message = node?.message || {};
        const metadata = message?.metadata || {};
        const sdk = metadata?.chatgpt_sdk || {};
        const responseMetadata = pickFirstObject(
            sdk?.response_metadata,
            sdk?.responseMetadata,
            metadata?.response_metadata,
            metadata?.responseMetadata,
        );

View on GitHub (pinned to 49907e53dc)

Solutions

Verify the request URL/conversation id matches the one used to fetch the payload (fix off-by-one id selection in the caller).
Bypass cache or include the conversation id in the cache key so stale payloads are never returned.
Log both ids on mismatch and re-fetch the conversation before extraction.
If the mismatch is expected (e.g. server-assigned ids), only pass expectedConversationId when strict verification is required.

Example fix

// before
const data = cache.get(anyConversationId) || await fetchConversation(id);
return extractDeepResearchFromConversationPayload(data, { expectedConversationId: id });
// after
const data = await fetchConversation(id); // cache key: `conv:${id}`
return extractDeepResearchFromConversationPayload(data, { expectedConversationId: id });

Defensive patterns

Strategy: validation

Validate before calling

const payloadId = String(payload?.conversation_id || payload?.conversationId || payload?.id || '').trim();
if (expectedId && payloadId && payloadId !== expectedId) {
  throw new Error(`stale/mismatched payload: ${payloadId} != ${expectedId}`);
}

Type guard

const matchesConversation = (p, id) =>
  !!p && typeof p === 'object' && String(p.conversation_id || p.conversationId || p.id || '').trim() === id;

Try / catch

try {
  return extractDeepResearchFromConversationPayload(payload, { expectedConversationId: id });
} catch (err) {
  if (String(err.message).includes('id mismatch')) {
    const fresh = await fetchConversation(id); // bypass cache and retry once
    return extractDeepResearchFromConversationPayload(fresh, { expectedConversationId: id });
  }
  throw err;
}

Prevention

Key caches by conversation id so stale payloads can't be served
Confirm the id you compare against is the conversation id, not a message id
Re-fetch the conversation when a mismatch is detected instead of reusing cached data
Log both expected and actual ids whenever this error fires

When it happens

Trigger: extractDeepResearchFromConversationPayload(payload, {expectedConversationId: 'X'}) receives a payload whose trimmed conversation_id/conversationId/id is a different non-empty string 'Y'.

Common situations: Caching layer returning a previous conversation's payload for a new id; backend reusing/redirecting to another conversation; client comparing against the wrong id field (e.g. message id vs conversation id); race where the conversation id changed after regeneration.

Related errors

AI-assisted analysis of jackwener/OpenCLI@49907e53dc (2026-08-29). Data as JSON: /api/errors/61051735cc02637f. Report an issue: GitHub.