PERFSCALE-4192: Jira mcp integration

README.md

@@ -79,7 +79,7 @@ BugZooka supports two complementary modes for monitoring Slack channels that can
    # Run with both polling AND socket mode
    make run ARGS="--product openshift --ci prow --enable-socket-mode"
    ```
-   
+
    **Socket Mode Requirements:**
    - An app-level token (`xapp-*`) must be configured as `SLACK_APP_TOKEN`
    - Socket Mode must be enabled in your Slack app settings
@@ -140,6 +140,7 @@ GENERIC_INFERENCE_URL="YOUR_INFERENCE_ENDPOINT"
 GENERIC_INFERENCE_TOKEN="YOUR_INFERENCE_TOKEN"
 GENERIC_MODEL="YOUR_INFERENCE_MODEL"
 ```
+
 **Note**: Please make sure to provide details for all the mandatory attributes and for the product that is intended to be used for testing along with fallback (i.e. GENERIC details) to handle failover use-cases.
 
 
@@ -315,6 +316,7 @@ BugZooka has a dependency on [orion-mcp service](https://github.com/jtaleric/ori
 export QUAY_CRED='<base64 encoded pull secret>'
 export BUGZOOKA_IMAGE='<bugzooka image tag>'
 export BUGZOOKA_NAMESPACE='<your namespace>'
+export JIRA_MCP_IMAGE='<jira mcp server image>'
 make deploy
 
 # Cleanup resources

bugzooka/analysis/prompts.py

@@ -110,7 +110,7 @@
 - No emojis in tables
 - Separate each config section with 80 equals signs.
 
-**Remember:** 
+**Remember:**
 - The tools provide percentage changes - use them as provided
 - CHECK thresholds (5% and 10%) before categorizing
 - SORT by absolute percentage change (highest first) - this is mandatory
@@ -127,3 +127,12 @@
 Beginning analysis now.
 """,
 }
+# Jira tool prompt - used when Jira MCP tools are available
+JIRA_TOOL_PROMPT = {
+    "system": (
+        "\n\nIMPORTANT: You have access to JIRA search tools. After analyzing the error, "
+        "ALWAYS search for related issues in JIRA using the search_jira_issues tool with the OCPBUGS project. "
+        "Extract key error terms, component names, or operators from the log summary to search for similar issues. "
+        "Include the top 3 most relevant JIRA issues in your final response under a 'Related JIRA Issues' section."
+    ),
+}

bugzooka/integrations/gemini_client.py

@@ -12,6 +12,7 @@
     INFERENCE_MAX_TOOL_ITERATIONS,
 )
 from bugzooka.integrations.inference import InferenceAPIUnavailableError
+from bugzooka.analysis.prompts import JIRA_TOOL_PROMPT
 
 
 logger = logging.getLogger(__name__)
@@ -44,7 +45,7 @@ def __init__(self, api_key=None, base_url=None, verify_ssl=None, timeout=None):
         # Timeout configuration
         if timeout is None:
             timeout = float(os.getenv("GEMINI_TIMEOUT", "60.0"))
-        
+
         logger.debug("Gemini client timeout set to %.1f seconds", timeout)
 
         # SSL verification configuration
@@ -79,19 +80,23 @@ def chat_completions_create(self, messages, model="gemini-2.0-flash", **kwargs):
         """
         try:
             logger.debug("Calling Gemini API: %s, Model=%s", self.base_url, model)
-            
+
             response = self.client.chat.completions.create(
                 model=model, messages=messages, **kwargs
             )
-            
+
             # Log token usage information
-            if hasattr(response, 'usage') and response.usage:
+            if hasattr(response, "usage") and response.usage:
                 usage = response.usage
-                logger.info("📊 Token usage - Prompt: %d, Completion: %d, Total: %d",
-                           usage.prompt_tokens, usage.completion_tokens, usage.total_tokens)
+                logger.info(
+                    "📊 Token usage - Prompt: %d, Completion: %d, Total: %d",
+                    usage.prompt_tokens,
+                    usage.completion_tokens,
+                    usage.total_tokens,
+                )
             else:
                 logger.debug("No usage information available in response")
-            
+
             logger.debug("Gemini API call successful")
             return response
         except Exception as e:
@@ -145,10 +150,10 @@ async def execute_tool_call(tool_name, tool_args, available_tools):
         logger.debug("Tool arguments: %s", json.dumps(tool_args, indent=2))
 
         # Check if the tool is async (has coroutine attribute or ainvoke method)
-        if hasattr(tool, 'coroutine') and tool.coroutine:
+        if hasattr(tool, "coroutine") and tool.coroutine:
             # MCP tools have a coroutine attribute
             result = await tool.ainvoke(tool_args)
-        elif hasattr(tool, 'ainvoke'):
+        elif hasattr(tool, "ainvoke"):
             # Some tools have ainvoke method
             result = await tool.ainvoke(tool_args)
         else:
@@ -158,19 +163,23 @@ async def execute_tool_call(tool_name, tool_args, available_tools):
         # Log result
         result_str = str(result)
         result_length = len(result_str)
-        
+
         # Check for empty or minimal results
         if not result_str or result_str.strip() in ["", "null", "None", "{}", "[]"]:
             logger.warning("⚠️ Tool %s returned empty or null result", tool_name)
         elif len(result_str.strip()) < 50:
-            logger.warning("⚠️ Tool %s returned small result (%d chars): %s", 
-                          tool_name, result_length, result_str)
+            logger.warning(
+                "⚠️ Tool %s returned small result (%d chars): %s",
+                tool_name,
+                result_length,
+                result_str,
+            )
         else:
             logger.info("✅ Tool %s completed (%d chars)", tool_name, result_length)
-            
+
         # Log full output at DEBUG level
         logger.debug("Tool %s output: %s", tool_name, result_str)
-        
+
         return result_str
     except Exception as e:
         error_msg = f"Error executing tool '{tool_name}': {str(e)}"
@@ -181,20 +190,17 @@ async def execute_tool_call(tool_name, tool_args, available_tools):
 
 
 async def analyze_with_gemini_agentic(
-    messages: list,
-    tools=None,
-    model="gemini-2.0-flash",
-    max_iterations=None
+    messages: list, tools=None, model="gemini-2.0-flash", max_iterations=None
 ):
     """
     Generic agentic loop for Gemini with tool calling support.
-    
+
     This function implements the agentic pattern where Gemini can iteratively:
     1. Analyze the current context
     2. Decide to call tools if needed
     3. Process tool results
     4. Generate final answer
-    
+
     :param messages: List of message dictionaries (system, user, assistant prompts)
     :param tools: List of LangChain tools available for Gemini to call (optional)
     :param model: Gemini model to use (default: gemini-2.0-flash)
@@ -203,18 +209,21 @@ async def analyze_with_gemini_agentic(
     """
     if max_iterations is None:
         max_iterations = INFERENCE_MAX_TOOL_ITERATIONS
-    
+
     try:
         gemini_client = GeminiClient()
-        
+
         # Convert LangChain tools to OpenAI format if provided
         openai_tools = None
         if tools:
             openai_tools = convert_langchain_tools_to_openai_format(tools)
             tool_names = [t["function"]["name"] for t in openai_tools]
-            logger.info("Starting Gemini analysis with %d tools: %s",
-                       len(openai_tools), ", ".join(tool_names))
-
+            logger.info(
+                "Starting Gemini analysis with %d tools: %s",
+                len(openai_tools),
+                ", ".join(tool_names),
+            )
+
         logger.debug("Starting agentic loop with %d messages", len(messages))
 
         # Tool calling loop - iterate until we get a final answer or hit max iterations
@@ -240,44 +249,51 @@ async def analyze_with_gemini_agentic(
             response_message = response.choices[0].message
 
             # Check if Gemini wants to call tools
-            tool_calls = getattr(response_message, 'tool_calls', None)
+            tool_calls = getattr(response_message, "tool_calls", None)
 
             if not tool_calls:
                 # No tool calls - we have the final answer
                 content = response_message.content
                 if content:
                     logger.info("Analysis complete after %d iteration(s)", iteration)
-                    logger.debug("Response: %s", content[:200] + "..." if len(content) > 200 else content)
+                    logger.debug(
+                        "Response: %s",
+                        content[:200] + "..." if len(content) > 200 else content,
+                    )
                 else:
                     logger.warning("Gemini returned None content, using empty string")
                     content = ""
                 return content
 
             # Gemini wants to call tools - execute them
             tool_names_called = [tc.function.name for tc in tool_calls]
-            logger.info("Calling %d tool(s): %s", len(tool_calls), ", ".join(tool_names_called))
+            logger.info(
+                "Calling %d tool(s): %s", len(tool_calls), ", ".join(tool_names_called)
+            )
 
             # Add the assistant's message with tool calls to conversation
-            messages.append({
-                "role": "assistant",
-                "content": response_message.content or "",
-                "tool_calls": [
-                    {
-                        "id": tc.id,
-                        "type": "function",
-                        "function": {
-                            "name": tc.function.name,
-                            "arguments": tc.function.arguments
+            messages.append(
+                {
+                    "role": "assistant",
+                    "content": response_message.content or "",
+                    "tool_calls": [
+                        {
+                            "id": tc.id,
+                            "type": "function",
+                            "function": {
+                                "name": tc.function.name,
+                                "arguments": tc.function.arguments,
+                            },
                         }
-                    }
-                    for tc in tool_calls
-                ]
-            })
+                        for tc in tool_calls
+                    ],
+                }
+            )
 
             # Execute each tool call and add results to messages
             for tool_call in tool_calls:
                 function_name = tool_call.function.name
-                
+
                 try:
                     function_args = json.loads(tool_call.function.arguments)
                 except json.JSONDecodeError as e:
@@ -286,25 +302,27 @@ async def analyze_with_gemini_agentic(
                 else:
                     # Execute the tool (await since it's now async)
                     function_result = await execute_tool_call(
-                        function_name,
-                        function_args,
-                        tools
+                        function_name, function_args, tools
                     )
 
                 # Add tool result to messages
-                messages.append({
-                    "role": "tool",
-                    "tool_call_id": tool_call.id,
-                    "name": function_name,
-                    "content": function_result
-                })
+                messages.append(
+                    {
+                        "role": "tool",
+                        "tool_call_id": tool_call.id,
+                        "name": function_name,
+                        "content": function_result,
+                    }
+                )
 
             # Continue loop to let Gemini process tool results
 
         # If we hit max iterations without a final answer
-        logger.warning("Reached maximum iterations (%d) without final answer", max_iterations)
+        logger.warning(
+            "Reached maximum iterations (%d) without final answer", max_iterations
+        )
         return "Analysis incomplete: Maximum tool calling iterations reached. Please try again with a simpler query."
-        
+
     except Exception as e:
         logger.error("Error in Gemini agentic loop: %s", str(e), exc_info=True)
         raise InferenceAPIUnavailableError(
@@ -318,7 +336,7 @@ async def analyze_log_with_gemini(
     error_summary: str,
     model="gemini-2.0-flash",
     tools=None,
-    max_iterations=None
+    max_iterations=None,
 ):
     """
     Analyzes log summaries using Gemini API with product-specific prompts and optional tool calling.
@@ -333,7 +351,7 @@ async def analyze_log_with_gemini(
     """
     try:
         logger.info("Starting log analysis for product: %s", product)
-        
+
         prompt_config = product_config["prompt"][product]
         try:
             formatted_content = prompt_config["user"].format(
@@ -342,20 +360,26 @@ async def analyze_log_with_gemini(
         except KeyError:
             formatted_content = prompt_config["user"].format(summary=error_summary)
 
-        logger.debug("Error summary: %s", error_summary[:150] + "..." if len(error_summary) > 150 else error_summary)
+        logger.debug(
+            "Error summary: %s",
+            error_summary[:150] + "..." if len(error_summary) > 150 else error_summary,
+        )
+
+        # Append Jira prompt if Jira MCP tools are available
+        system_prompt = prompt_config["system"]
+        if tools and any(getattr(t, "name", "") == "search_jira_issues" for t in tools):
+            logger.info("Jira MCP tools detected - injecting Jira prompt")
+            system_prompt += JIRA_TOOL_PROMPT["system"]
 
         messages = [
-            {"role": "system", "content": prompt_config["system"]},
+            {"role": "system", "content": system_prompt},
             {"role": "user", "content": formatted_content},
             {"role": "assistant", "content": prompt_config["assistant"]},
         ]
 
         # Use the generic agentic loop
         return await analyze_with_gemini_agentic(
-            messages=messages,
-            tools=tools,
-            model=model,
-            max_iterations=max_iterations
+            messages=messages, tools=tools, model=model, max_iterations=max_iterations
         )
 
     except Exception as e:

kustomize/base/configmap-mcp-config.yaml

@@ -12,6 +12,10 @@ data:
         "orion_mcp_server": {
           "url": "http://orion-mcp.orion-mcp:3030/mcp",
           "transport": "streamable_http"
+        },
+        "jira_mcp_server": {
+          "url": "http://jira-mcp.${BUGZOOKA_NAMESPACE}:3031/mcp",
+          "transport": "streamable_http"
         }
       }
     }

kustomize/base/configmap-prompts.yaml

@@ -8,11 +8,11 @@ metadata:
 data:
   prompts.json: |
     {
-      "OPENSHIFT_PROMPT": {
-        "system": "You are an expert in OpenShift, Kubernetes, and cloud infrastructure. Your task is to analyze logs and summaries related to OpenShift environments. Given a log summary, identify the root cause, potential fixes, and affected components. Be as consise as possible (under 5000 characters), but precise and avoid generic troubleshooting steps. Prioritize OpenShift-specific debugging techniques. Keep in mind that the cluster is ephemeral and is destroyed after the build is complete, but all relevant logs and metrics are available. Use markdown formatting for the output with only one level of bullet points, do not use bold text except for the headers.",
-        "user": "Here is the log summary from an OpenShift environment:\n\n{summary}\n\nBased on this summary, provide a structured breakdown of:\n- The OpenShift component likely affected (e.g., etcd, kube-apiserver, ingress, SDN, Machine API)\n- The probable root cause\n- Steps to verify the issue further\n- Suggested resolution, including OpenShift-specific commands or configurations.",
-        "assistant": "**Affected Component:** <Identified component>\n\n**Probable Root Cause:** <Describe why this issue might be occurring>\n\n**Verification Steps:**\n- <Step 1>\n- <Step 2>\n- <Step 3>\n\n**Suggested Resolution:**\n- <OpenShift CLI commands>\n- <Relevant OpenShift configurations>"
-      },
+        "OPENSHIFT_PROMPT": {
+    "system": "You are an expert in OpenShift, Kubernetes, and cloud infrastructure. Your task is to analyze logs and summaries related to OpenShift environments. Given a log summary, identify the root cause, potential fixes, and affected components. Be as consise as possible (under 5000 characters), but precise and avoid generic troubleshooting steps. Prioritize OpenShift-specific debugging techniques. Keep in mind that the cluster is ephemeral and is destroyed after the build is complete, but all relevant logs and metrics are available. Use markdown formatting for the output with only one level of bullet points, do not use bold text except for the headers.",
+    "user": "Here is the log summary from an OpenShift environment:\n\n{summary}\n\nBased on this summary, provide a structured breakdown of:\n- The OpenShift component likely affected (e.g., etcd, kube-apiserver, ingress, SDN, Machine API)\n- The probable root cause\n- Steps to verify the issue further\n- Suggested resolution, including OpenShift-specific commands or configurations",
+    "assistant": "**Affected Component:** <Identified component>\n\n**Probable Root Cause:** <Describe why this issue might be occurring>\n\n**Verification Steps:**\n- <Step 1>\n- <Step 2>\n- <Step 3>\n\n**Suggested Resolution:**\n- <OpenShift CLI commands>\n- <Relevant OpenShift configurations>"
+    },
       "ANSIBLE_PROMPT": {
         "system": "You are an expert in Ansible automation, playbook debugging, and infrastructure as code (IaC). Your task is to analyze log summaries related to Ansible execution, playbook failures, and task errors. Given a log summary, identify the root cause, affected tasks, and potential fixes. Prioritize Ansible-specific debugging techniques over generic troubleshooting.",
         "user": "Here is the log summary from an Ansible execution:\n\n{summary}\n\nBased on this summary, provide a structured breakdown of:\n- The failed Ansible task and module involved\n- The probable root cause\n- Steps to reproduce or verify the issue\n- Suggested resolution, including relevant playbook changes or command-line fixes.",