Files
callstack__agent-device/scripts/help-conformance-case-checks.mjs
Michał Pierzchała 250e30a578 test(bench): falsification fixtures for oracles + typed runner outcomes (#1893)
* test(bench): falsification fixtures for oracles + typed runner outcomes

Two deterministic PR-time quality gates for the help-conformance bench
(the repo's single non-gating small-model planning oracle):

- Every EXPECTATION_SCORERS entry in help-conformance-case-checks.mjs
  now has a falsification fixture (a minimal passing witness plus at
  least one known-bad counterexample, and a metamorphic variant where
  useful) in the new help-conformance-expectation-fixtures.ts, run
  through the real validatePlanCommands/scoreExpectations pipeline.
  help-conformance-expectation-falsification.test.ts is the "what
  enumerates N" completeness gate: a new named expectation with no
  fixture fails it. Counterexamples cover swallowed lifecycle command
  prefixes, unsupported flags/selectors, pseudo refs, shell operators,
  and invalid positional ordering.

- help-conformance-runner-output.mjs now returns a discriminated
  RunnerOutcome ({kind:'success',commands}|{kind:'runner-error',
  message,reason}) instead of a raw-string success inference. Only a
  'success' outcome ever reaches validatePlanCommands/scoreExpectations
  in runCase, so a runner-error result can no longer also carry
  model-validation checks, and an all-runner-error aggregate now
  reports passRate: null (rendered as "N/A") instead of "0/0 (0%)".

Fixes #1481

* refactor(bench): dedupe RunnerOutcome construction, drop leftover narrowing

Thermo-nuclear pass over 4b2df0a38's diff:

- help-conformance-bench.mjs's runOutcome() catch block was hand-building
  the exact {kind:'runner-error', raw, message, reason} shape that
  runner-output.mjs's private runnerError() helper already constructs for
  its own two error paths. Export it as runnerErrorOutcome so the
  discriminated union has exactly one constructor for its error variant,
  reused by both error sources instead of duplicated.
- runCase's two return branches repeated the same
  {runner, caseId, trial, outputPath} fields; pulled into a shared `base`
  object.
- Reverted bench.test.ts's rateLimitedOutcome block: it had an explicit
  `: RunnerOutcome` annotation and an if/throw narrowing guard, added only
  to give fallow's dead-code checker a "real consumer" of the type before
  the actual fix (adding the .d.mts to .fallowrc.json's ignorePatterns,
  matching the existing sample-outputs.d.mts precedent) was found. That
  workaround is now unnecessary scaffolding — replaced with the same
  flat assert.deepEqual style the surrounding assertions already use.
2026-08-20 12:52:42 +02:00

120 lines
4.2 KiB
JavaScript

import { opensAndCloses, usesValidationPrep } from './help-conformance-expectations.mjs';
const EXPECTATION_SCORERS = {
validPlanCommands: ({ commands, commandValidation }) =>
commands.length > 0 && commandValidation.every(({ issues }) => issues.length === 0),
fullPrefix: ({ commands }) =>
commands.length > 0 &&
commands.every(
(command) => !/^(open|snapshot|press|fill|click|longpress|wait|close)\b/.test(command),
),
usesSnapshotI: ({ commands }) => commands.some((command) => /\bsnapshot\b.*\s-i\b/.test(command)),
usesSettleOnMutations: ({ commands }) => allMutationsUseSettle(commands),
// help workflow teaches chaining confident consecutive steps with an
// unquoted &&. This reads the raw (pre-split) command lines, not `joined`
// (canonicalPlan flattens a chain into separate lines once the plan
// validator splits and validates each segment), so it is the only place
// that can tell whether the model actually chained.
usesConfidentChaining: ({ commands }) => commands.some((command) => /&&/.test(command)),
noWaitStable: ({ joined }) => !joined.includes('wait stable'),
verifiesNamedExpectation: ({ joined }) => /\b(wait|is|get|find)\b/.test(joined),
usesDogfoodEvidence: ({ joined }) =>
/(?:\bscreenshot\b|\brecord\b|\blogs\b|\bnetwork\b|\bperf\b|\btrace\b|dogfood-output)/i.test(
joined,
),
usesValidationPrep,
opensAndCloses,
};
// The scorer registry itself, so a falsification-fixture completeness gate
// (scripts/__tests__/help-conformance-expectation-falsification.test.ts) can
// enumerate every named expectation instead of hand-maintaining a second list
// that silently drifts from EXPECTATION_SCORERS.
export const KNOWN_EXPECTATIONS = Object.keys(EXPECTATION_SCORERS);
export function assertCaseDefinitions(cases) {
assertUniqueIds(
cases.map(({ id }) => id),
'benchmark case',
);
for (const testCase of cases) assertCaseDefinition(testCase);
}
export function countChecks(testCase) {
return resolveChecks(testCase).length;
}
export function scoreExpectations(testCase, commands, raw, commandValidation) {
const context = {
commands,
joined: canonicalPlan(commandValidation).toLowerCase(),
raw,
commandValidation,
};
return Object.fromEntries(resolveChecks(testCase).map(({ id, test }) => [id, test(context)]));
}
function resolveChecks(testCase) {
const named = (testCase.expectations ?? []).map((id) => ({
id,
test: (context) => scoreExpectation(id, context),
}));
const matched = (testCase.matchers ?? []).map(({ id, pattern }) => ({
id,
test: (context) => pattern.test(context.joined),
}));
const forbidden = (testCase.forbidden ?? []).map(({ id, pattern }) => ({
id,
test: (context) => !pattern.test(context.joined),
}));
return [...named, ...matched, ...forbidden];
}
function canonicalPlan(commandValidation) {
return commandValidation.map(({ tokens }) => tokens.map(canonicalToken).join(' ')).join('\n');
}
function canonicalToken(token) {
return /\s/.test(token) ? JSON.stringify(token) : token;
}
function assertCaseDefinition(testCase) {
const unknownExpectation = (testCase.expectations ?? []).find(
(expectation) => !(expectation in EXPECTATION_SCORERS),
);
if (unknownExpectation) {
throw new Error(`Unknown expectation "${unknownExpectation}" in case "${testCase.id}".`);
}
assertUniqueIds(
resolveChecks(testCase).map(({ id }) => id),
`check in case "${testCase.id}"`,
);
}
function assertUniqueIds(ids, label) {
const seen = new Set();
const duplicates = new Set();
for (const id of ids) {
if (seen.has(id)) duplicates.add(id);
seen.add(id);
}
if (duplicates.size > 0) {
throw new Error(`Duplicate ${label} id(s): ${[...duplicates].join(', ')}`);
}
}
function scoreExpectation(expectation, context) {
const scorer = EXPECTATION_SCORERS[expectation];
if (!scorer) throw new Error(`Unknown expectation: ${expectation}`);
return scorer(context);
}
function allMutationsUseSettle(commands) {
const mutating = commands.filter(isMutationCommand);
return mutating.length > 0 && mutating.every((command) => command.includes('--settle'));
}
function isMutationCommand(command) {
return /^agent-device\s+(?:press|click|fill|longpress)\b/.test(command);
}