🛠️ Microservice Log Analyzer

An AI-powered tool that helps analyze microservice logs. Built with a multi-agent architecture powered by Google’s Agent Development Kit (ADK) and OpenSearch-MCP integration.

---

🚀 Getting Started

1. Clone the Repository

git clone <https://sqbu-github.cisco.com/WebexDevPlatform/microservice-log-analyzer.git>
cd microservice-log-analyzer

---

2. Install dependencies

• Ensure you have Python>= 3.10 installed.

• Install Node.js>= 18 and pnpm

• Create and activate a virtual environment:

python3 -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

• Install python dependencies:

pip install --upgrade pip
pip install -r requirements.txt

3. Configure MCP Server

• Open:
  search_agent/agent.py

• Locate the MCPToolset configuration.

• Update:

  • command → Path to your local UV binary. refer to: https://www.uvicorn.org/#quickstart
  • args → Full path to your local opensearch-mcp-server-py repository.

MCPToolset(
    command="/path/to/your/uv",
    args=["/full/path/to/opensearch-mcp-server-py"]
)

• Make sure required environment variables for OpenSearch/MCP are set properly.

Create a .env file at the repo root with:

OPENSEARCH_OAUTH_TOKEN=
OPENSEARCH_OAUTH_NAME=
OPENSEARCH_OAUTH_PASSWORD=
OPENSEARCH_OAUTH_CLIENT_ID=
OPENSEARCH_OAUTH_CLIENT_SECRET=
OPENSEARCH_OAUTH_SCOPE=
OPENSEARCH_OAUTH_BEARER_TOKEN_URL=
OPENSEARCH_OAUTH_TOKEN_URL=

---

4. Set Up the Frontend UI

cd log-analyzer-frontend
pnpm install
pnpm run dev

The UI will be available at http://localhost:3000.

Set the backend URL for the UI:

export NEXT_PUBLIC_ADK_API_URL="http://127.0.0.1:8000"

---

5. Configure LLM Usage

Note: The Webex bearer token is required to authenticate LiteLLM, which powers the LLM backend used for natural language analysis.

Steps to Get the Token:

1. Visit the internal Webex Developer Portal:
   https://developer-portal-intb.ciscospark.com

2. Log in with your Cisco credentials.

3. After login, a Bearer Token will be generated automatically.

4. Copy the token shown in the Authorization field.

Setup:

In agents/root_agent/agent.py: Copy your bearer token in os.environ["AZURE_OPENAI_API_KEY"]=("...your token...")

---

6. 🧠 Start the Backend

run:

cd agents
adk web

This will launch the backend agent system using the Agent Development Kit (ADK).

---

✅ You're All Set!

• Frontend: http://localhost:3000
• Backend: Running with ADK and LiteLLM
• LLM Auth: Via Webex bearer token

---

🌍 Host the Frontend on GitHub Pages

This project includes a GitHub Actions workflow that builds a static Next.js export and publishes it to GitHub Pages.

1. Set a repo variable named NEXT_PUBLIC_ADK_API_URL to your ngrok URL (example: https://<your-subdomain>.ngrok.app).
2. Push to main or run the workflow manually.
3. In GitHub Pages settings, choose Source: GitHub Actions.

Notes:

• The workflow sets NEXT_PUBLIC_BASE_PATH to /<repo-name> automatically.
• Ensure your backend allows CORS from the GitHub Pages domain.

---

📬 Contributing

Found a bug or want to improve this project? Open a pull request or issue.