[FEATURE] Connectors to customize generated APIs and code
- Dominant language
- Python
- Stars
- 13
- Forks
- 1
- PR merge metrics
- No merged PRs in 30d
Description
# Add Support for Connectors That Consider Customer Context When Generating Code Using RAG
**Description:**
The Industry Toolkit will be enhanced with connectors that can take customer-specific context, such as coding standards, existing codebases, and architectural patterns, into account when generating code. This feature will leverage Retrieval-Augmented Generation (RAG) to provide tailored code generation, ensuring that the output aligns with the customer's established guidelines and practices.
### Feature Requirements:
- Implement connectors that utilize RAG to retrieve customer-specific context (coding standards, existing repositories, etc.) and use that information to guide code generation.
- Ensure the connectors can adapt to a variety of customer environments, including different coding styles, frameworks, and design patterns.
- Provide seamless integration with customer repositories and other context sources to retrieve relevant information for RAG.
- Generate code that aligns with customer-specific practices, ensuring it fits within their existing infrastructure.
- Write comprehensive tests to validate that the generated code adheres to customer context.
- Include documentation on how to configure the connectors and use the customer context feature.
### Acceptance Criteria:
- [ ] Connectors that use RAG to incorporate customer context in code generation are implemented.
- [ ] Code generated by the connectors aligns with customer-specific coding standards and architecture.
- [ ] Integration with customer repositories and context sources is functional.
- [ ] Tests are written and passing to validate customer-specific code generation.
- [ ] README and documentation are updated with instructions on configuring and using customer context for code generation.
Contributor guide
Research direction
Start with the README and existing documentation, then locate the current code-generation and connector entry points before defining how customer repositories and other context sources are configured. Done means implementing the RAG-based connectors, validating customer-specific output with comprehensive tests, and documenting configuration and usage.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai, devtools
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100