I built an app that gives me bad advice - and it's in Jira
summary
Builds an AI agent inside Jira using Forge CLI and Atlassian Rovo tooling and deploys it via forge deploy and forge install to a development environment then connects it with Jira automation to comment on new issues. The app is a ReactJS Forge app with manifest.yaml, uses forge tunnel for live testing and Forge storage for state, and successfully posts intentionally silly comments to tickets but a prompt that explicitly asks for harmful bad advice is blocked by platform checks.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 3:18Disclaimers
- 7:15Atlassian Forge platform
- 17:31Forge app from scratch
- 29:36Deploying and testing
- 36:44Custom prompts
- 44:10Jira automations
- 51:45Live demo
- 57:04Advanced use cases
- 58:28Q&A
related talks
24mDevelopers and AI Enemy or Frenemy Leveraging AI and Atlassian to empower developers
32mStop Chat-and-Hope and Start Building Better Apps with Agents
38mAI for Software Development: A Reality Check
37mPatterns of Application Development Using AI
32mOptimizing App Development with Generative AI
17mFrom Silos to Synergy: How Platform Engineering Bridges Teams with AI
48mThe Room of Requirements for Developers: Scaling AI Magic with Internal Portals
28mGenerative AI-powered app development