By name: testing description: Guides ccusage Rust and Node tests. Use when adding or fixing cargo tests, Node test files, CLI snapshots, Claude model pricing, LiteLLM compatibility, or fixture-backed tests.
ccusage Testing
Use the tdd skill for logic changes, Red-Green-Refactor workflow, focused
runner commands, Rust test syntax, and general test readability rules. This
skill owns ccusage-specific Rust and Node test rules: fixtures, adapter
coverage, pricing/model behavior, CLI output, schema artifacts, and package
tooling.
Shared Rules
- Prefer behavior-focused tests over schema-shape tests unless schema normalization itself is the behavior.
- Avoid
try/catchin tests for expected failures. Use native result, throwing, rejecting, or explicit failure assertions for the test framework. - Avoid
ifbranches inside test bodies. Split behaviors into separate tests or table-driven cases. - Tests do not need to be DRY. Prefer repeated, explicit setup when duplication makes each behavior easier to read.
- Do not hoist one-off values out of tests when sharing them hides the behavior.
- Skipped local-data smoke tests are acceptable when real user log directories catch schema drift, but they must not fail on clean CI machines.
Rust Tests
For repo-wide validation, prefer the just recipe because it runs Node and Rust tests together:
direnv exec . just test
Put Rust unit tests near the module they exercise with #[cfg(test)] mod tests.
When splitting a large module, move its tests with the code instead of leaving
unrelated tests in main.rs.
Read references/rust.md for fixture-backed parser, path discovery, SQLite
loading, dedupe, aggregation, pricing, CLI output parity, model, and readability
rules. For focused cargo commands, Red-Green-Refactor examples, Rust test
attributes, and syntax examples, read ../tdd/references/rust.md.
Node Tests
Read references/node-test.md for TypeScript package launcher, schema artifact,
benchmark script, docs/package tooling, and filesystem fixture rules.