Uploaded November 2026 | Updated November 2026, 3 days ago
Learn how to troubleshoot failed CI/CD builds with an AI agent that retrieves build runs, checks build status, analyzes execution logs, and recommends a corrective action. This end-to-end tutorial uses OCI DevOps, OCI Logging, an MCP server, and Oracle Private Agent Factory to investigate a real build failure.
Build a CI/CD pipeline for a Python order-service application, run automated PyTest unit tests, and intentionally introduce a discount-calculation defect. The AI troubleshooting agent retrieves build evidence through MCP tools, identifies the failed tests and probable root cause, and recommends the fix. It helps developers understand the failure without automatically changing application code.
You will learn how to create an OCI DevOps project and repository, configure a managed build pipeline and build logging, connect MCP tools to an AI agent workflow, validate a successful build, and diagnose a failed build from operational evidence.
Chapters:
0:00 Architecture for AI-assisted CI/CD troubleshooting
2:11 Create the DevOps project and code repository
5:10 Configure the Python application, tests, and build specification
7:42 Create the managed build pipeline
9:28 Enable build execution logging
10:43 Configure MCP tools and the AI troubleshooting agent
14:31 Validate a successful build
16:25 Diagnose a failed build and verify MCP tool calls
Project files and setup instructions:
github.com/pmccormick01/tutorial-paf-troubleshooting-agent
Explore more Code Innovate videos:
youtube.com/playlist?list=PLPIzp-E1msrZMCfSHbKgLK3KWsNM9JB9a
#codeinnovate #CICD #DevOps #AIAgents #MCP
Learn how to troubleshoot failed CI/CD builds with an AI agent that retrieves build runs, checks build status, analyzes execution logs, and recommends a corrective action. This end-to-end tutorial uses OCI DevOps, OCI Logging, an MCP server, and Oracle Private Agent Factory to investigate a real build failure.
Build a CI/CD pipeline for a Python order-service application, run automated PyTest unit tests, and intentionally introduce a discount-calculation defect. The AI troubleshooting agent retrieves build evidence through MCP tools, identifies the failed tests and probable root cause, and recommends the fix. It helps developers understand the failure without automatically changing application code.
You will learn how to create an OCI DevOps project and repository, configure a managed build pipeline and build logging, connect MCP tools to an AI agent workflow, validate a successful build, and diagnose a failed build from operational evidence.
Chapters:
0:00 Architecture for AI-assisted CI/CD troubleshooting
2:11 Create the DevOps project and code repository
5:10 Configure the Python application, tests, and build specification
7:42 Create the managed build pipeline
9:28 Enable build execution logging
10:43 Configure MCP tools and the AI troubleshooting agent
14:31 Validate a successful build
16:25 Diagnose a failed build and verify MCP tool calls
Project files and setup instructions:
github.com/pmccormick01/tutorial-paf-troubleshooting-agent
Explore more Code Innovate videos:
youtube.com/playlist?list=PLPIzp-E1msrZMCfSHbKgLK3KWsNM9JB9a
#codeinnovate #CICD #DevOps #AIAgents #MCP










