TencentCloud/TencentDB-Agent-Memory · error · LlmResolveError

llm.provider=proxy 需要 llm.baseUrl 指向 context_proxy 根 URL (如

Error message

llm.provider=proxy 需要 llm.baseUrl 指向 context_proxy 根 URL (如 http://127.0.0.1:8096)

What it means

When llm.provider is set to "proxy", resolveEffectiveLlmConfig requires llm.baseUrl to point at the context_proxy root URL, because all LLM calls are rewritten to <baseUrl>/proxy/<iid>/v1. If baseUrl is missing/empty, the gateway cannot build the proxy endpoint and throws LlmResolveError.

Source

Thrown at MemoryCore/src/gateway/llm-resolver.ts:55

 * @param memorySystemUser 已解析的 memory 系统用户配置（可选，仅 provider=proxy 且
 *                         useMemorySystemUserKey=true 时使用）
 *
 * @throws {LlmResolveError} provider=proxy 但缺失必要配置时
 */
export function resolveEffectiveLlmConfig(
  llm: StandaloneLLMConfig,
  instanceId: string | undefined,
  memorySystemUser: MemorySystemUserConfig | undefined,
): StandaloneLLMConfig {
  const provider = llm.provider ?? "openai";
  if (provider !== "proxy") {
    // 默认路径：原样返回，行为完全等价改造前
    return llm;
  }

  // provider=proxy 的校验
  if (!llm.baseUrl) {
    throw new LlmResolveError(
      "llm.provider=proxy 需要 llm.baseUrl 指向 context_proxy 根 URL (如 http://127.0.0.1:8096)",
    );
  }
  if (!instanceId || !instanceId.trim()) {
    throw new LlmResolveError(
      "llm.provider=proxy 需要 instanceId，但 core 当前 instanceId 为空 —— " +
      "service 模式下确保请求带 x-tdai-service-id，standalone 模式下确保 yaml 有 instanceId",
    );
  }

  const useSystemUserKey = llm.proxy?.useMemorySystemUserKey ?? true;
  let effectiveApiKey = llm.apiKey;
  if (useSystemUserKey) {
    if (!memorySystemUser) {
      throw new LlmResolveError(
        "llm.provider=proxy 且 llm.proxy.useMemorySystemUserKey=true 需要 " +
        "metadata.systemUser.memory 完整配置（userId + userKey），当前缺失",
      );

View on GitHub (pinned to 3efcd317b8)

Solutions

1. Set llm.baseUrl to the context_proxy root, e.g. baseUrl: http://127.0.0.1:8096 (no trailing path like /v1 — the resolver appends /proxy/<iid>/v1).
2. Ensure the context_proxy service is deployed and its address is reachable from MemoryCore.
3. If you don't intend to use the proxy, set llm.provider back to the direct provider (e.g. openai) with its own baseUrl/apiKey.

Example fix

// before
llm:
  provider: proxy
  apiKey: sk-...
// after
llm:
  provider: proxy
  baseUrl: http://127.0.0.1:8096
  apiKey: sk-...

if (llmConfig.provider === 'proxy' && (!llmConfig.baseUrl || !llmConfig.baseUrl.trim())) {
  throw new Error('llm.provider=proxy requires llm.baseUrl, e.g. http://127.0.0.1:8096');
}

function hasProxyBaseUrl(llm: { provider: string; baseUrl?: string }): llm is typeof llm & { baseUrl: string } {
  return llm.provider !== 'proxy' || (typeof llm.baseUrl === 'string' && llm.baseUrl.length > 0);
}

try {
  gateway.start();
} catch (err) {
  if (err instanceof LlmResolveError && err.message.includes('llm.baseUrl')) {
    console.error('Set llm.baseUrl to the context_proxy root URL');
  }
  throw err;
}

When it happens

Trigger: Configuring llm: { provider: proxy } in the gateway yaml (or runtime LLM config) without llm.baseUrl, or with baseUrl: "" after a migration from direct provider mode.

Common situations: Switching provider from "openai"/direct to "proxy" but deleting the old baseUrl; copying a config snippet that omitted baseUrl; forgetting that provider=proxy ignores provider-specific hosts and requires the context_proxy service address (e.g. http://127.0.0.1:8096).

Related errors

• llm.provider=proxy 需要 llm.baseUrl 指向 context_proxy 根 URL
• llm.provider=proxy 且 useMemorySystemUserKey=true 需要 metadata
• cos.generationLogRetentionDays must be a non-negative intege
• llm.provider=proxy 需要 instanceId，但 core 当前 instanceId 为空 ——
• llm.provider=proxy 且 llm.proxy.useMemorySystemUserKey=true 需

AI-assisted analysis of TencentCloud/TencentDB-Agent-Memory@3efcd317b8 (2026-09-01). Data as JSON: /api/errors/56a1cc8f1fafdf35. Report an issue: GitHub.