Thank you for your interest in contributing to HyperAgent!
We use GitHub Flow:
main → stable, always deployable
└── feat/* → new features (feat/sse-transport)
└── fix/* → bug fixes (fix/cache-expiry)
└── docs/* → documentation updates
└── refactor/* → code refactoring
└── test/* → test improvements
# For features
git checkout -b feat/your-feature-name
# For bug fixes
git checkout -b fix/issue-number-description
# For docs
git checkout -b docs/what-youre-documenting- Write tests for new functionality
- Ensure all tests pass:
cargo test --workspace - Format code:
cargo fmt --all - Check lints:
cargo clippy --workspace
Use conventional commits:
feat: add SSE transport for MCP server
fix: resolve cache expiry timing issue
docs: add approvals guide
refactor: simplify executor logic
test: add integration tests for engine
- Push your branch:
git push origin feat/your-feature - Open PR against
main - Fill out the PR template
- Wait for CI to pass
cargo fmt --allcargo clippy --workspace --all-targets -- -D warnings# Run all tests
cargo test --workspace
# Run specific crate tests
cargo test -p cortex-mcp
# Run with output
cargo test --workspace -- --nocapturecrates/
├── agents/ # Core agent functionality
├── core/ # Shared types and traits
├── providers/ # LLM backends (OpenAI, Anthropic, etc.)
├── mcp/ # Model Context Protocol
├── cache/ # Caching layer
├── resilience/ # Circuit breaker, retry, etc.
├── workflow/ # YAML workflow DSL
└── ...
See ROADMAP.md for current priorities:
- Testing - Integration tests with MockBackend
- MCP - SSE transport, Agent-as-MCP-tool
- Documentation - Guides for backends, approvals
Open an issue or start a discussion!