feat: add traceable scoped AIOps diagnosis

2026-07-04 22:57:28 +08:00
parent 246c99b954
commit 23ee05c7c3
32 changed files with 1179 additions and 25 deletions
@@ -10,11 +10,12 @@ import com.superbiz.agent.agent.tool.InternalDocsTools;
 import com.superbiz.agent.agent.tool.QueryLogsTools;
 import com.superbiz.agent.agent.tool.QueryMetricsTools;
 import com.superbiz.agent.domain.entity.AgentStep;
-import com.superbiz.agent.domain.entity.AgentStep;
 import com.superbiz.agent.domain.entity.DiagnosisSession;
+import com.superbiz.agent.dto.AIOpsRequest;
 import com.superbiz.agent.hook.AgentLoggingHook;
 import com.superbiz.agent.repository.AgentStepRepository;
 import com.superbiz.agent.repository.DiagnosisSessionRepository;
+import com.superbiz.agent.repository.ToolInvocationRepository;
 import com.superbiz.agent.util.SessionContextHolder;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
@@ -62,6 +63,9 @@ public class AiOpsService {
    @Autowired
    private AgentStepRepository agentStepRepository;

+    @Autowired
+    private ToolInvocationRepository toolInvocationRepository;
+
    /**
     * 执行 AI Ops 告警分析流程
     *
@@ -71,22 +75,22 @@ public class AiOpsService {
     * @throws GraphRunnerException 如果 Agent 执行失败
     */
    public Optional<OverAllState> executeAiOpsAnalysis(ChatModel chatModel, ToolCallback[] toolCallbacks) throws GraphRunnerException {
+        return executeAiOpsAnalysis(chatModel, toolCallbacks, null, resolveSessionId(null));
+    }
+
+    public Optional<OverAllState> executeAiOpsAnalysis(ChatModel chatModel, ToolCallback[] toolCallbacks,
+                                                       AIOpsRequest request, String sessionId) throws GraphRunnerException {
        logger.info("开始执行 AI Ops 多 Agent 协作流程");

-        String sessionId = UUID.randomUUID().toString().substring(0, 8);
+        String resolvedSessionId = isBlank(sessionId) ? resolveSessionId(request) : sessionId.trim();
        long startTime = System.currentTimeMillis();

-        // 创建诊断会话
-        DiagnosisSession session = DiagnosisSession.builder()
-                .sessionId(sessionId)
-                .query("AI Ops 告警分析")
-                .status("RUNNING")
-                .agentFlow("AI_OPS")
-                .build();
+        // 创建或更新诊断会话
+        DiagnosisSession session = startDiagnosisSession(resolvedSessionId, request);
        diagnosisSessionRepository.save(session);

        // 设置 ThreadLocal 上下文（LookupKnowledgeTool 通过此获取 sessionId）
-        SessionContextHolder.setSessionId(sessionId);
+        SessionContextHolder.setSessionId(resolvedSessionId);

        try {
            // 构建 Planner 和 Executor Agent（每个 Agent 各自带 Hook）
@@ -102,7 +106,7 @@ public class AiOpsService {
                    .subAgents(List.of(plannerAgent, executorAgent))
                    .build();

-            String taskPrompt = "你是企业级 SRE，接到了自动化告警排查任务。请结合工具调用，执行**规划→执行→再规划**的闭环，并最终按照固定模板输出《告警分析报告》。禁止编造虚假数据，如连续多次查询失败需诚实反馈无法完成的原因。";
+            String taskPrompt = buildTaskPrompt(request);

            logger.info("调用 Supervisor Agent 开始编排...");

@@ -158,6 +162,87 @@ public class AiOpsService {
        }
    }

+    public String resolveSessionId(AIOpsRequest request) {
+        if (request != null && !isBlank(request.getSessionId())) {
+            return request.getSessionId().trim();
+        }
+        return UUID.randomUUID().toString();
+    }
+
+    public void persistFinalReport(String sessionId, String finalReport) {
+        if (isBlank(sessionId) || isBlank(finalReport)) {
+            return;
+        }
+        diagnosisSessionRepository.findBySessionId(sessionId.trim()).ifPresent(session -> {
+            session.setAnswer(finalReport);
+            diagnosisSessionRepository.save(session);
+        });
+    }
+
+    String buildQuerySummary(AIOpsRequest request) {
+        if (request == null) {
+            return "AI Ops 告警分析";
+        }
+
+        StringBuilder summary = new StringBuilder("AI Ops 告警分析");
+        appendField(summary, "告警", request.getAlertName());
+        appendField(summary, "服务", request.getService());
+        appendField(summary, "等级", request.getSeverity());
+        appendField(summary, "时间范围", request.getTimeRange());
+        appendField(summary, "描述", request.getDescription());
+        appendField(summary, "请求", request.getUserRequest());
+        return summary.toString();
+    }
+
+    boolean hasAlertPayload(AIOpsRequest request) {
+        if (request == null) {
+            return false;
+        }
+        return !isBlank(request.getAlertName())
+                || !isBlank(request.getService())
+                || !isBlank(request.getSeverity())
+                || !isBlank(request.getDescription())
+                || !isBlank(request.getTimeRange());
+    }
+
+    String buildTaskPrompt(AIOpsRequest request) {
+        StringBuilder prompt = new StringBuilder();
+        prompt.append("你是企业级 SRE，接到了自动化告警排查任务。请结合工具调用，执行**规划→执行→再规划**的闭环，并最终按照固定模板输出《告警分析报告》。禁止编造虚假数据，如连续多次查询失败需诚实反馈无法完成的原因。");
+        prompt.append("\n\n本次告警输入：\n");
+        prompt.append(buildQuerySummary(request));
+        if (hasAlertPayload(request)) {
+            prompt.append("\n\nAIOps scope mode: PAYLOAD_TARGETED\n");
+            prompt.append("- The request includes an alert payload. Treat the supplied alert payload as the primary and only main diagnosis target.\n");
+            prompt.append("- The final report must focus on the supplied alert fields such as alertName, service, severity, description, and timeRange.\n");
+            prompt.append("- You may call queryPrometheusAlerts only to verify whether the supplied alert is still active or to identify related risk/context.\n");
+            prompt.append("- If queryPrometheusAlerts returns unrelated active alerts, do not create full root-cause or remediation sections for them.\n");
+            prompt.append("- Mention unrelated active alerts only briefly in a Related Risk section when they help explain the supplied alert.\n");
+        } else {
+            prompt.append("\n\nAIOps scope mode: AUTO_DISCOVERY\n");
+            prompt.append("- The request does not include alert payload fields. First call queryPrometheusAlerts to discover current active/firing alerts.\n");
+            prompt.append("- Prefer P0/P1 alerts or the longest-running firing alerts, then diagnose one or more alerts based on severity and evidence.\n");
+            prompt.append("- Use metrics, logs, and knowledge-base evidence before producing the final alert analysis report.\n");
+        }
+        return prompt.toString();
+    }
+
+    private DiagnosisSession startDiagnosisSession(String sessionId, AIOpsRequest request) {
+        DiagnosisSession session = diagnosisSessionRepository.findBySessionId(sessionId)
+                .orElseGet(() -> DiagnosisSession.builder()
+                        .sessionId(sessionId)
+                        .agentFlow("AI_OPS")
+                        .build());
+        session.setQuery(buildQuerySummary(request));
+        session.setStatus("RUNNING");
+        session.setAgentFlow("AI_OPS");
+        session.setAnswer(null);
+        session.setTotalDurationMs(null);
+        session.setTotalTokenCount(null);
+        session.setStepCount(null);
+        session.setToolCallCount(null);
+        return session;
+    }
+
    /**
     * 构建 Planner Agent
     */
@@ -205,25 +290,33 @@ public class AiOpsService {
        }
    }

-    /** 从 agent_step 汇总指标回填 diagnosis_session */
+    /** 从 agent_step 和 tool_invocation 汇总指标回填 diagnosis_session */
    private void backfillSessionMetrics(DiagnosisSession session) {
        try {
            List<AgentStep> steps = agentStepRepository.findBySessionIdOrderByStepIndex(session.getSessionId());
-            if (steps.isEmpty()) return;

            int totalTokens = 0;
            int stepCount = 0;
-            int toolCallCount = 0;
            for (AgentStep s : steps) {
                stepCount++;
                if (s.getTokenCount() != null) totalTokens += s.getTokenCount();
-                if (Boolean.TRUE.equals(s.getHasToolCall())) toolCallCount++;
            }
+            long toolCallCount = toolInvocationRepository.countBySessionId(session.getSessionId());
            session.setTotalTokenCount(totalTokens);
            session.setStepCount(stepCount);
-            session.setToolCallCount(toolCallCount);
+            session.setToolCallCount(Math.toIntExact(toolCallCount));
        } catch (Exception e) {
            logger.warn("回填会话指标失败: sessionId={}", session.getSessionId(), e);
        }
    }
+
+    private void appendField(StringBuilder builder, String label, String value) {
+        if (!isBlank(value)) {
+            builder.append("\n- ").append(label).append(": ").append(value.trim());
+        }
+    }
+
+    private boolean isBlank(String value) {
+        return value == null || value.trim().isEmpty();
+    }
 }