feat(prose-tests): the framework — cases, worlds, runner, skill#545
Open
leeovery wants to merge 2 commits into
Open
feat(prose-tests): the framework — cases, worlds, runner, skill#545leeovery wants to merge 2 commits into
leeovery wants to merge 2 commits into
Conversation
Stage 2 of design/prose-tests.md. The deterministic layer in code: case parser + corpus validation, fixture recipes under a frozen clock with golden snapshots (rebuild-compare in npm test), the world builder (installed-layout skills, hermetic git, keyword-only knowledge store), and the runner whose `prompt` command is the P4 boundary — walker prompts never carry expects. The /prose-test dev skill owns the model layer: Sonnet walks, Opus confirms failures, quoted evidence required. First fixture (base) proven byte-deterministic; two smoke cases exercised the full pipe end-to-end. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Review finding (Lee): coordinate-phrased cases (step numbers, numbered heading anchors) break on cosmetic renumbering — failure for the wrong reason. Anchors become substring fragments (includes-matched, still red token-free when the heading vanishes), the smoke cases are rewritten purely behaviourally, and the authoring rule lands in the README. Prose test hooks considered and parked in the design log: the walker binds semantically, so shipped-prose plumbing isn't earning its keep — revisit on evidence at the failure harvest. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
design/prose-tests.md): case parser + corpus validation (tests/prose/lib/cases.cjs), fixture recipes run under a frozen-clock preload with committed golden snapshots (lib/fixtures.cjs,lib/fake-clock.cjs), the world builder (lib/world.cjs— installed-layout skills, hermetic git, keyword-only knowledge store re-derived at materialise), and the runner CLI (run.cjs: list · select · world · prompt · grade · snap · verify · destroy).run.cjs promptand never containexpect:claims;graderuns state assertions deterministically via the world's own engine and emits routing claims for the grader agent only.npm test: corpus validation (parse, scoped paths, anchors, assertion grammar) and fixture golden checks (every snapshot rebuilds byte-identical from its recipe — drift lands as a reviewable snapshot diff, never silently). Plus the/prose-testdev skill (Sonnet walks, Opus confirms failures, quoted evidence required) and thebasefixture with two smoke cases.Test plan
npm testgreen (1683 tests, includes both new suites; base fixture rebuild ~2.9s).node tests/prose/run.cjs verifyrun twice, byte-identical both times.engine bootrun in-world (knowledge ready, zero migrations), deterministic grade passed, walker prompt verified expect-free, world destroyed.🤖 Generated with Claude Code