chore: update basic setup & readme

Signed-off-by: Tomas Weiss <tomas.weiss2@gmail.com>

Author: tomkis

README.md

@@ -14,7 +14,7 @@
 A multi-agent system for creating well-structured GitHub issues using the [BeeAI Framework](https://github.com/i-am-bee/beeai-framework). This system coordinates between specialized agents to draft issues, check for duplicates, and create final GitHub issues with customizable templates. The system is consumable via the [Agent Stack](https://github.com/i-am-bee/agentstack) (UI and CLI) through the A2A protocol.
 
 ✅ Multi-agent workflow  
-🔄 Real-time trajectory tracking  
+🔄 Real-time trajectory tracking
 📝 Customizable issue templates
 
 ## What is GitHub Issue Creator?
@@ -70,6 +70,21 @@ uv run server
 
 The server will start on `http://127.0.0.1:8000` and register the GitHub Issue Creator agent that coordinates the complete workflow.
 
+### Running in Agent Stack
+
+You can also easily start the agent in [Agent Stack](https://agentstack.beeai.dev/).
+
+```bash
+# Install the GH issue creator
+agentstack add ghcr.io/i-am-bee/github-issue-creator/github-issue-creator
+
+# Setup the repo
+agentstack env add "GitHub Issue Creator" GITHUB_REPOSITORY=username/reponame
+
+# Setup the PAT
+agentstack env add "GitHub Issue Creator" GITHUB_PAT=github_pat_XXX
+```
+
 ## Security Considerations
 
 > [!WARNING]

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "github-issue-creator"
-version = "0.1.0"
+version = "0.2.0"
 readme = "README.md"
 authors = [
     {name = "Matous Havlena", email = "matous.havlena@ibm.com"}

src/github_issue_creator/utils/config.py

@@ -1,9 +1,20 @@
+import os
+from beeai_framework.backend import ChatModel
 from dotenv import load_dotenv
+
 from beeai_framework.adapters.agentstack.backend.chat import AgentStackChatModel
 
 load_dotenv()
 
-llm = AgentStackChatModel(preferred_models=["openai:gpt-4o"])
+
+if os.getenv("API_KEY") is not None:
+    print("Using API key from environment variable")
+    model = os.getenv("MODEL", "openai:gpt-5-mini")
+    llm = ChatModel.from_name(model, {"api_key": os.getenv("API_KEY")})
+else:
+    print("Using LLM from Agent Stack")
+    model = os.getenv("MODEL", "openai:gpt-4o")
+    llm = AgentStackChatModel(preferred_models=["openai:gpt-4o"])
 
 # Import after load_dotenv to ensure env vars are loaded
 from github_issue_creator.tools.session_manager import SessionManager