alibaba/page-agent · error · InvokeError

UNKNOWN

UNKNOWN

Error message

HTTP ${response.status}: ${errorMessage}

What it means

The endpoint returned a non-OK HTTP status that is not 401/403, 429, or 5xx — typically 400/404/405/422 class errors. The status code and the provider's error message are embedded, with the parsed body attached.

Source

Thrown at packages/llms/src/OpenAIClient.ts:122

					`Authentication failed: ${errorMessage}`,
					errorData
				)
			}
			if (response.status === 429) {
				throw new InvokeError(
					InvokeErrorTypes.RATE_LIMIT,
					`Rate limit exceeded: ${errorMessage}`,
					errorData
				)
			}
			if (response.status >= 500) {
				throw new InvokeError(
					InvokeErrorTypes.SERVER_ERROR,
					`Server error: ${errorMessage}`,
					errorData
				)
			}
			throw new InvokeError(
				InvokeErrorTypes.UNKNOWN,
				`HTTP ${response.status}: ${errorMessage}`,
				errorData
			)
		}

		// 4. Parse and validate response
		let data: any
		try {
			data = await response.json()
		} catch (error) {
			if ((error as any)?.name === 'AbortError') throw error
			throw new InvokeError(
				InvokeErrorTypes.INVALID_RESPONSE,
				'Response body is not valid JSON',
				error
			)
		}

View on GitHub (pinned to d02db1ee7c)

Solutions

1. Read the embedded message and errorData — providers are specific about what's wrong
2. Verify the model name is valid for this endpoint (GET <baseURL>/models)
3. Check baseURL includes the correct path prefix (e.g. /v1)
4. If 400 on tool calls, simplify tool parameter schemas and re-test

Example fix

// before
new OpenAIClient({ baseURL: 'https://api.openai.com', model: 'gpt4-turbo' }) // 404s

// after
new OpenAIClient({ baseURL: 'https://api.openai.com/v1', model: 'gpt-4-turbo' })

// smoke-test config before use
const res = await fetch(`${cfg.baseURL}/models`, { headers: { Authorization: `Bearer ${cfg.apiKey}` } })
if (!res.ok) throw new Error(`Endpoint config problem: HTTP ${res.status}`)

const isUnknownHttpError = (e: unknown): e is InvokeError =>
  e instanceof InvokeError && e.code === 'UNKNOWN' && /^HTTP \d+/.test(e.message)

try {
  await client.invoke(req)
} catch (e) {
  if (isUnknownHttpError(e)) {
    console.error('Inspect status & body:', e.message, e.details) // includes errorData
    throw e
  }
  throw e
}

When it happens

Trigger: invoke() with a malformed payload the provider rejects with 400 (bad model name, invalid message format, unknown tool schema); 404 from a wrong baseURL path (missing /v1) or nonexistent model/deployment; 422 validation errors from compatible providers.

Common situations: Model name typo or model not available to this endpoint; baseURL missing the /v1 path segment so /chat/completions 404s; tool definitions with unsupported JSON schema features; provider-specific required params missing; version drift between client payload shape and provider API.

Understand the failure class

Background: "API error: {status}" and "HTTP 401/403/404/429/5xx" errors: non-2xx HTTP responses explained — this error's family across 27 libraries.

Related errors

• CONFIG_ERROR
• NETWORK_ERROR
• AUTH_ERROR
• RATE_LIMIT
• SERVER_ERROR

AI-assisted analysis of alibaba/page-agent@d02db1ee7c (2026-08-28). Data as JSON: /api/errors/8802bc85fe5bc234. Report an issue: GitHub.