Skip to content

Switchyard lets LLM applications route traffic across models and providers while preserving native OpenAI and Anthropic API compatibility - enabling flexible model selection, benchmarking, and cost/performance optimization.

Rust

doc(AGENTS.md): Allow panic in tests (#209)

Graham Kinge4e5f47c39a9
diff --git a/AGENTS.md b/AGENTS.md
index ab5b7febc..4b5190309 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -351,7 +351,7 @@ uvicorn.run(build_switchyard_app(switchyard), port=4000)
- **Type hints**: throughout. `py.typed` marker present; mypy runs strict.
- **Async**: async-only. If you need sync, use `asyncio.run()`.
- **Testing**: `respx` for HTTP mocking, `pytest-mock` for general mocking. `asyncio_mode = "auto"` — no `@pytest.mark.asyncio` needed.
-- **Rust**: do not use `.expect()` in Rust source or tests. Propagate errors with `?`,
+- **Rust**: Panicking calls such as `panic!()`, `unwrap()` and `.except()` are not allowed in production source code. Propagate errors with `?`. They are allowed in tests.
return typed errors, or match explicitly in tests so failures stay intentional and visible.
- **Comments**: For Rust changes, add concise comments for module/file intent, public structs/enums,
public methods, private helpers with non-obvious behavior, and tests that encode important