Skip to content

test(tutor-quality): keep only the coverage lower bound for strategy cases (R-AUT-1) #156

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ttbombadil merged 1 commit into from
Oct 3, 2026
+3 −2
Merged

test(tutor-quality): keep only the coverage lower bound for strategy cases (R-AUT-1) #156

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions tests/server/services/tutor/evaluation/anchor-corpus.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,15 +43,16 @@ function validSource(): TutorQualityCorpusSource {
}

describe("Tutor Quality anchor corpus contract", () => {
it("contains two judge-enabled strategy cases for each learning phase", () => {
it("contains at least two judge-enabled strategy cases for each learning phase", () => {
const source = parseYaml(readFileSync(fileURLToPath(new URL("../../../../../evals/tutor-quality/anchor-corpus.yaml", import.meta.url)), "utf8")) as TutorQualityCorpusSource;
const strategyCases = source.scenarios.filter(({ id }) => id.startsWith("strategy-"));
const counts = Object.fromEntries(["LEARN", "DEEPEN", "EXPAND"].map((phase) => [
phase,
strategyCases.filter(({ expected }) => expected?.learningPhase === phase).length,
]));

expect(Object.values(counts).every((count) => count >= 2 && count <= 3)).toBe(true);
// R-AUT-1: a lower bound for coverage, never a case count; new cases are corpus data only.
expect(Object.values(counts).every((count) => count >= 2)).toBe(true);
expect(strategyCases.every(({ turns, judge, expected }) => (
turns.some(({ kind }) => kind === "dialog")
&& expected?.learningPhase !== undefined
Expand Down
Loading