GustAI provides a VS Code extension that creates project‑folder scoped AI agents which automatically ingest local code, configuration files, and style guidelines. These context‑aware agents generate and refactor code with precision, inherit knowledge across hierarchical scopes, and can be chained or have their memory tuned to build custom automation pipelines for developers.
Funding
Funding not disclosed
Founders
Product
Problem
Software developers often rely on generic AI code assistants that lack awareness of project-specific conventions, configurations, and dependencies, leading to inaccurate suggestions and frequent manual corrections. This context mismatch reduces productivity and increases the risk of introducing errors into codebases.
Solution
GustAI offers a VS Code extension that creates context-aware AI agents scoped to individual project folders. Each agent learns the local coding standards, configuration files, and project structure, enabling precise code generation, refactoring, and updates. Agents inherit insights from parent scopes, maintaining consistency across related modules while preventing drift. The platform plans to integrate vector databases, live APIs, and external services to enrich the agents' knowledge base, ensuring they operate with up-to-date information. By chaining agents and tuning their memory, developers can build complex, automated workflows that function as collaborative team members.
Target Audience
Primary users are software developers and engineering teams seeking AI assistance that respects project-specific conventions and can be integrated directly into their development environment.
Features
- Project-folder scoped agents that automatically ingest local code, configs, and style guidelines
- Hierarchical scope inheritance allowing shared knowledge across related modules
- VS Code integration for in-editor prompting, code generation, and refactoring
- Planned connectivity to vector embeddings, APIs, and external services for enriched context
- Ability to chain multiple agents and adjust memory settings to create custom automation pipelines
- Beta release with controlled tester access to gather feedback and refine functionality