Skip to content

test: restore out-of-range answer coverage for all four questions - #61

Merged
BunsDev merged 1 commit into
mainfrom
test/out-of-range-all-questions
Oct 7, 2026
Merged

BunsDev merged 1 commit into
mainfrom
test/out-of-range-all-questions

Conversation

@BunsDev

@BunsDev BunsDev commented Oct 7, 2026

Copy link
Copy Markdown
Contributor

What changed

Restores a playground test that the phase 1 extraction (#32) did not carry over: decision table: out-of-range probabilities and confidence never permit from typesafe-playground tests/proposal-review.test.ts at 408ebb55086fb9b6d517dc5a11aa1c404117f699 (#41 head; tree-identical to merge 6fe5967).

The existing tests/answer-invariants.test.ts case covered addresses_task only and asserted the verdict, not the named miss. The restored test runs all 4 question ids × {probability, confidence} × {−0.01, 1.01, −2, 2} (32 cases). For each case it asserts proposal_only and checks that the reason contains <id>: no usable answer. It also asserts permit at p = 1/0. The only adaptation is the harness's review-envelope helper.

Mutation check: if decide()'s confidence > 1 bound is removed, the existing invariant test still passes, because the contradiction check catches the case with a different reason. The restored test fails. So this closes a real gap in the regression net.

Verdict impact

None. This PR changes tests only. No decision-table row, threshold, question wording or version, fixture, or mock verdict changes.

Verification

  • pnpm typecheck
  • pnpm test: 346 pass, 0 fail (pnpm verify: boundaries, typecheck, tests, build, secrets, all passing)
  • Commits are signed
  • No provider call, credential, execution path, or new verdict added
  • Decision table and questions unchanged

🤖 Generated with Claude Code

The phase 1 extraction did not carry over the playground test "decision
table: out-of-range probabilities and confidence never permit"
(typesafe-playground tests/proposal-review.test.ts at 408ebb5). The
existing invariant test covered addresses_task only and asserted the
verdict, not the named miss. This restores the full 4 ids x {probability,
confidence} x {-0.01, 1.01, -2, 2} grid and asserts each reason names
"<id>: no usable answer", plus permit at p = 1/0.

Test-only. No verdict, decision-table row, threshold, or fixture changes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@vercel

vercel Bot commented Oct 7, 2026 •

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
jev-harness Ready Ready Preview Oct 7, 2026 12:51pm UTC

Request Review

@BunsDev
BunsDev merged commit 90e1cf4 into main Oct 7, 2026
3 checks passed
@BunsDev
BunsDev deleted the test/out-of-range-all-questions branch October 7, 2026 15:22

This branch was successfully deployed

1 active deployment
Preview — 06826fdf Deployed Oct 7, 2026 by vercel[bot]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant