Blog post about kooq agent

content/blog/first-koog-agent-automating-developer-workflows.md

@@ -0,0 +1,117 @@
+---
+author: karol
+tags:
+  - AI
+  - Koog
+date: 2025-07-17T08:50:18.274Z
+meaningfullyUpdatedAt: 2025-06-17T14:02:33.275Z
+title: Your First Koog Agent Automating Developer Workflows
+layout: post
+image: 
+hidden: false
+comments: true
+published: true
+language: en
+---
+
+# Your First Koog Agent: Automating Developer Workflows
+
+Most of the teams working in IT relay on tools like Jira and Slack to manage work and communicate. 
+But what if you could automate routine tasks—like checking a Jira ticket status and posting an update to Slack,
+using a single natural language command? With Koog Agents, you can.
+
+In this post, I'll guide you how to build your first Koog Agent that connects Jira and Slack, using annotation-based tools and the power of single-run agents.
+
+---
+
+## Why Automate Developer Workflows?
+
+Developers often switch between tools to check ticket statuses and update teammates and stakeholders.
+Automating these steps saves time and reduces context switching, letting you focus on what matters: building great software.
+
+---
+
+## Introducing Koog Agents
+
+Koog Agents are lightweight, annotation-driven components that let you expose business logic as tools, accessible via natural language. 
+With single-run agents, you can execute a sequence of actions in response to a single command.
+
+---
+
+## Defining Tools with Annotations
+
+Let's look at a `TicketUpdateToolSet` that exposes two tools: one for Jira, one for Slack.
+
+```kotlin
+@LLMDescription("A set of tools for updating Jira tickets and sending messages to Slack.")
+class TicketUpdateToolSet(
+    private val jiraApi: JiraApi,
+    private val slackApi: SlackApi,
+): ToolSet {
+
+    @Tool
+    @LLMDescription("Retrieves the status of a Jira ticket by its key.")
+    suspend fun jiraTool(
+        @LLMDescription("The key of the Jira ticket, e.g. BA-1")
+        ticketKey: String,
+    ): String {
+        return jiraApi.getTicketStatus(ticketKey).let {
+            "The status of the Jira ticket $ticketKey is: $it"
+        }
+    }
+
+    @Tool
+    @LLMDescription("Sends a message to a Slack channel.")
+    suspend fun slackTool(
+        @LLMDescription("The name of the Slack channel where the message should be sent")
+        channel: String,
+        @LLMDescription("The text of the message to be sent")
+        text: String,
+    ): String {
+        val isMessageSent = slackApi.sendMessage(channel, text)
+        return if (isMessageSent) "Message was sent." else "Message could not be sent."
+    }
+}
+```
+
+With just a few annotations, you expose your business logic as tools that Koog Agents can use.
+
+---
+
+## How It Works
+
+1. **Natural Language Command:**  
+   The developer says:  
+   _"Check the status of ticket BA-1 and post an update to #dev-team on Slack."_
+
+2. **Agent Orchestration:**  
+   The Koog Agent parses the command, calls `jiraTool` to get the ticket status, then calls `slackTool` to post the update.
+
+3. **Single-Run Simplicity:**  
+   The entire workflow runs in a single agent execution — no manual steps required.
+
+---
+
+## Benefits
+
+- **Rapid Automation:** Expose new tools with simple annotations.
+- **Natural Language:** Trigger complex workflows with a single command.
+- **Integration:** Connect your existing APIs and services.
+
+---
+
+## Get Started
+
+1. Define your tools using the `@Tool` and `@LLMDescription` annotations.
+2. Register your toolset with a Koog Agent.
+3. Start automating your developer workflows!
+
+---
+
+Koog Agents make it easy to bridge the gap between your tools and your team. 
+With annotation-based tools and single-run agents, you can automate repetitive tasks and keep your team in sync—using just your words.
+
+###
+Links to useful documentation:
+https://github.com/JetBrains/koog
+https://docs.koog.ai/annotation-based-tools/#implement-the-toolset-interface