## Inspiration
Modern software teams make hundreds of code changes every day. While automated testing catches many technical issues, important business rules are often hidden inside old support tickets, bug reports, documentation, and developer comments. These undocumented rules are easily forgotten, leading to production incidents, financial losses, and compliance risks.
RuleGuard AI was built to solve this problem by using multiple AI agents that automatically discover hidden business rules, evaluate software changes, verify supporting evidence, and determine whether a release is safe before deployment.
What it does
RuleGuard AI is a multi-agent release intelligence platform that acts as an AI-powered release gate for software teams.
Instead of only checking code quality, it analyzes historical company knowledge and compares it against proposed software updates.
The workflow consists of four specialized AI agents:
• Change Agent – Understands the proposed software modification. • Rule Agent – Discovers hidden business rules from historical company records. • Test Agent – Validates whether the proposed change violates discovered rules. • Safety Agent – Produces the final release recommendation with supporting evidence.
The platform provides explainable AI outputs by showing the discovered rules, evidence sources, reasoning, risk assessment, and a final Safe or Unsafe release verdict.
How we built it
The frontend was developed using React, TypeScript, and Vite.
The backend logic was implemented using Node.js and deployed as Netlify Serverless Functions, allowing the complete application to run entirely on Netlify without requiring dedicated backend hosting.
The multi-agent workflow was implemented using modular orchestration where each AI agent performs an independent responsibility before contributing to the final release decision.
Challenges we ran into
- Designing a realistic multi-agent workflow instead of a single AI response.
- Creating explainable outputs rather than simple Safe/Unsafe predictions.
- Migrating from a local Express backend to Netlify Serverless Functions.
- Ensuring the entire project could be deployed without paid cloud infrastructure.
Accomplishments that we're proud of
- Built a fully functional multi-agent software release analysis system.
- Successfully deployed the complete application on Netlify.
- Designed an explainable AI workflow with evidence-backed reasoning.
- Demonstrated both Safe and Unsafe release scenarios using real business-rule examples.
What we learned
This project reinforced the importance of explainable AI, agent orchestration, serverless deployment, and designing AI systems that assist human decision-making instead of replacing it.
What's next for RuleGuard AI
Future development will transform RuleGuard AI into an enterprise-ready release intelligence platform with:
- GitHub and GitLab integration
- Automatic PR analysis
- Jira and Azure DevOps support
- Retrieval-Augmented Generation (RAG) over enterprise documentation
- Real LLM-powered rule discovery
- Compliance and security policy validation
- Team dashboards and analytics
- CI/CD pipeline integration
- Enterprise deployment support
What it does
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for RuleGuard AI
Built With
- agents
- ai
- context
- css
- devops
- engineering
- express.js
- functions
- git
- github
- html5
- javascript
- llms
- management
- multi-agent
- netlify
- node.js
- openai
- prompt
- react
- release
- risk
- software
- typescript
- vite
Log in or sign up for Devpost to join the conversation.