{"record":{"id":"620e7819a8f12356","repo":"continuedev/continue","slug":"error-getting-prompt-mcp-connection-servername","errorCode":null,"errorMessage":"Error getting prompt: MCP Connection ${serverName} not found","messagePattern":"Error getting prompt: MCP Connection (.+?) not found","errorType":"exception","errorClass":"Error","httpStatus":null,"severity":"error","filePath":"core/context/mcp/MCPManagerSingleton.ts","lineNumber":195,"sourceCode":"  getStatuses(): (MCPServerStatus & { client: Client })[] {\n    return Array.from(this.connections.values()).map((connection) => ({\n      ...connection.getStatus(),\n      client: connection.client,\n    }));\n  }\n\n  setStatus(serverId: string, status: MCPServerStatus[\"status\"]) {\n    this.connections.get(serverId)!.status = status;\n  }\n\n  async getPrompt(\n    serverName: string,\n    promptName: string,\n    args: Record<string, string> = {},\n  ) {\n    const connection = this.connections.get(serverName);\n    if (!connection) {\n      throw new Error(\n        `Error getting prompt: MCP Connection ${serverName} not found`,\n      );\n    }\n    return await connection.client.getPrompt({\n      name: promptName,\n      arguments: args,\n    });\n  }\n}\n","sourceCodeStart":177,"sourceCodeEnd":205,"githubUrl":"https://github.com/continuedev/continue/blob/5522c6f44ca0ac3528b37244818fbfa39b5af470/core/context/mcp/MCPManagerSingleton.ts#L177-L205","documentation":"The anthropic-on-Vertex streaming branch received a non-2xx response; the response body is JSON-stringified into the message. Represents an upstream HTTP failure during a streaming chat request.","triggerScenarios":"401/403 auth or missing model access (anthropic model not enabled in Vertex Model Garden); 404 bad model id; 429 quota; 400 malformed converted payload.","commonSituations":"Claude model not subscribed/enabled in the GCP project; wrong region for the Anthropic endpoint; token expiry during long-lived streaming sessions; quota exhaustion under load.","solutions":["Parse the embedded JSON body for Google's specific error reason","Enable the Anthropic model in Model Garden and grant the service account access","Verify model id format claude-*@<version> and region","Add retry with exponential backoff for 429/5xx around the stream start"],"exampleFix":null,"handlingStrategy":"retry","validationCode":null,"typeGuard":null,"tryCatchPattern":"try { for await (const c of api.chatCompletionStream(body, signal)) {} } catch (e) { const msg = (e as Error).message; if (/429|5\\d\\d/.test(msg)) await retryWithBackoff(); else throw e; }","preventionTips":["Enable publisher models in Model Garden before streaming","Retry stream initiation (not mid-stream) on 429/5xx","Log the embedded JSON body for postmortems"],"tags":["vertexai","streaming","http-error","anthropic"],"backgroundTag":"upstream-api-http-error","analyzedSha":"5522c6f44ca0ac3528b37244818fbfa39b5af470","analyzedAt":"2026-08-27T11:28:54.683Z","schemaVersion":2},"datasetVersion":"2026-08-27T13:17:12.746Z"}