Uploaded November 2026 | Updated November 2026, 13 hours ago
In this episode of AI Coding with Dr. Chuck, I wake up at 4 AM with an idea for implementing LTI 1.3 Names and Role Provisioning Services (NRPS) in Tsugi. I can't get back to sleep, so I get up and start coding with Cursor AI.
Because it's four in the morning, this is another silent coding session. There is a spoken introduction and conclusion, but most of the video consists of screen recording and background music without narration.
The goal is to implement two important parts of LTI 1.3: OAuth 2.0 access token granting and the Names and Role Provisioning Service, which allows an LTI tool to retrieve course membership and role information from the learning platform.
Building on the LTI 1.3 Dynamic Registration, Deep Linking, and resource link launch work from previous episodes, we now have the architecture needed to add standards-based services.
In this session:
- Implementing LTI 1.3 Names and Role Provisioning Services (NRPS)
- Building the OAuth 2.0 access token granting process
- Connecting an LTI tool to Tsugi's membership service
- Testing the implementation using the Tsugi LMS interoperability test tool
- Debugging an access token failure
- Distinguishing token-granting problems from NRPS problems
- Using detailed protocol debugging output to identify and correct the issue
- Completing the implementation in two rounds of prompting
- Reviewing the implementation with CodeRabbit
- Tightening security-related details identified during code review
- Merging the completed changes into the master branch
The first attempt is surprisingly successful. The LTI launch works, but the Names and Roles request fails.
Initially, I suspect the NRPS implementation or possibly the test tool. After examining the debugging output, I realize the problem occurs earlier in the protocol: the tool isn't successfully obtaining an access token.
LTI services require a tool to request an access token before calling protected service endpoints. Once we identify and correct the token problem, the Names and Roles request works.
The existing Tsugi interoperability test tool is particularly useful here. I originally built it to help developers test LTI implementations across learning management systems. Its detailed debugging output makes it possible to diagnose the protocol failure and provide Cursor with the information needed to fix it.
After getting the service working, I submit the changes for CodeRabbit review. CodeRabbit identifies two areas worth tightening. We make those adjustments, receive a successful follow-up review, and merge the changes into master.
This continues my effort to build outbound LTI 1.3 platform support in Tsugi, using an architecture that supports registrations, deployments, messages, resource links, and services.
The next major step is Assignment and Grade Services (AGS), which will add the ability for external tools to interact with assignments, line items, and grades.
**About this video's format:** This is an intentional silent coding recording with a spoken introduction and conclusion. The middle portion has background music rather than narration, so you can follow the development process on screen.
Tsugi: tsugi.org
Python for Everybody: py4e.com
I'll be teaching Python for Everybody with AI beginning in January 2027. Visit py4e.com for more information about upcoming courses.
#Tsugi #LTI13 #NRPS #CursorAI #AICoding #SoftwareEngineering
In this episode of AI Coding with Dr. Chuck, I wake up at 4 AM with an idea for implementing LTI 1.3 Names and Role Provisioning Services (NRPS) in Tsugi. I can't get back to sleep, so I get up and start coding with Cursor AI.
Because it's four in the morning, this is another silent coding session. There is a spoken introduction and conclusion, but most of the video consists of screen recording and background music without narration.
The goal is to implement two important parts of LTI 1.3: OAuth 2.0 access token granting and the Names and Role Provisioning Service, which allows an LTI tool to retrieve course membership and role information from the learning platform.
Building on the LTI 1.3 Dynamic Registration, Deep Linking, and resource link launch work from previous episodes, we now have the architecture needed to add standards-based services.
In this session:
- Implementing LTI 1.3 Names and Role Provisioning Services (NRPS)
- Building the OAuth 2.0 access token granting process
- Connecting an LTI tool to Tsugi's membership service
- Testing the implementation using the Tsugi LMS interoperability test tool
- Debugging an access token failure
- Distinguishing token-granting problems from NRPS problems
- Using detailed protocol debugging output to identify and correct the issue
- Completing the implementation in two rounds of prompting
- Reviewing the implementation with CodeRabbit
- Tightening security-related details identified during code review
- Merging the completed changes into the master branch
The first attempt is surprisingly successful. The LTI launch works, but the Names and Roles request fails.
Initially, I suspect the NRPS implementation or possibly the test tool. After examining the debugging output, I realize the problem occurs earlier in the protocol: the tool isn't successfully obtaining an access token.
LTI services require a tool to request an access token before calling protected service endpoints. Once we identify and correct the token problem, the Names and Roles request works.
The existing Tsugi interoperability test tool is particularly useful here. I originally built it to help developers test LTI implementations across learning management systems. Its detailed debugging output makes it possible to diagnose the protocol failure and provide Cursor with the information needed to fix it.
After getting the service working, I submit the changes for CodeRabbit review. CodeRabbit identifies two areas worth tightening. We make those adjustments, receive a successful follow-up review, and merge the changes into master.
This continues my effort to build outbound LTI 1.3 platform support in Tsugi, using an architecture that supports registrations, deployments, messages, resource links, and services.
The next major step is Assignment and Grade Services (AGS), which will add the ability for external tools to interact with assignments, line items, and grades.
**About this video's format:** This is an intentional silent coding recording with a spoken introduction and conclusion. The middle portion has background music rather than narration, so you can follow the development process on screen.
Tsugi: tsugi.org
Python for Everybody: py4e.com
I'll be teaching Python for Everybody with AI beginning in January 2027. Visit py4e.com for more information about upcoming courses.
#Tsugi #LTI13 #NRPS #CursorAI #AICoding #SoftwareEngineering










