* feat(decision): add Clef and Jev executors on the System One choice API
* feat(decision): default Clef transport to clef-flash
* feat(decision): generalize to decision models, address review
* fix(decision): reject non-string decision choices
* fix(decision): skip empty select labels and blank destinations
An empty option label or whitespace-only destination offered as a choice fails at dispatch with INVALID_ARGUMENT and aborts the step instead of letting the model re-aim. Omit them at enumeration. Also qualify the uncertain-action docs: blocking stops the selected action, while earlier confident actions in the step have already run.
* feat(decision): add opt-in vision for Clef, Jev stays text-only
The executor stays text-only by default. With vision true and a transport that declares it, masked viewport pixels ride along as System One images data URLs. Clef declares vision; Jev never receives pixels. Withheld or tainted viewports fall back to semantic text.
* docs(decision): qualify vision cost to calls carrying pixels
* fix(decision): untainted pixel fixtures and honest skill pixels line
The test helper takes a tainted flag so the image-send tests use the only state the runner returns pixels in; tainted captures stay withheld observations. The skill no longer claims pixels are unavailable: screenshots may inform node choices with vision true, without adding point actions.
* bench(testbed): luna vs clef-flash harness and first numbers
Same todo flow under two configs: default agent on gpt-6-luna via OpenCodeX and decisionExecutor on clef-flash. Luna side passed in 23s with 6 model calls; clef-flash pending credentials. Includes the probe script, the docs section, and the flow screenshot.
* docs(decision): fill clef-flash benchmark row
Two no-cache runs both blocked uncertain on the first action (a25 at p=0.619, conf=0.387 against the default 0.9 gates), answering in about a second with 1,543 input tokens.
* docs(decision): benchmark aggregates over 5 runs per side
Luna completed 5/5 flows at ~14s and ~25k input tokens per run; clef-flash blocked uncertain 5/5 at ~1s and ~1.5k tokens per run.
* bench(testbed): parametrized flow, gate sweep, wire-shape finding
Params fix the missing type choices; gates 0.6/0.5 and 0.51/0.3 still block. Direct experiment shows hierarchical structured questions take clef-flash from 0.57 to 0.89 confidence on the same screen.
* feat(decision): hierarchical operation-then-target on structured transports
Structured transports (Clef, Jev) now decide the operation first and the target within it, with object instructions and element records. Flat endpoints keep the single string choice. Live benchmark: first operation 0.909/0.802, first target 0.940/0.775, up from 0.747/0.558 flat.
* docs(decision): compare luna vs clef-flash vs clef with measured cost
* feat(decision): run decisions through AI SDK evaluate, gates opt-in
decisionExecutor now accepts any AI SDK evaluation model that answers
choice questions, such as typeSafeAi.evaluationModel('jev-latest'), and
decides through experimental_evaluate (one choice question per decision,
no retries, so the call budget stays exact). ai is an optional peer that
loads only on that path. evaluationModel() exposes clef(), jev(), and
systemOne() transports to experimental_evaluate, which covers Clef until
it has an AI SDK provider.
The official TypeSafe provider rounds probabilities to two places, so a
long distribution can sum to 0.99: validation now allows half a unit per
rounded option, the AI SDK rule, and jev() declares the two places.
Probability and confidence gates are off by default and stay
configurable. Measuring clef-flash without them surfaced three executor
bugs, each fixed here:
- a lone target is dispatched without a one-option question, which the
choice API rejects with HTTP 422;
- acting decisions receive the step's action history, and the repeat
guard counts the same action on the same screen across the whole step,
so an A-B oscillation blocks instead of burning the call budget;
- the operation and target questions state that typed text is not saved
until it is submitted.
The structured path also checks the call budget before the target
request (review feedback).
Testbed todo flow, 5 no-cache runs each: clef-flash 5/5 (9.3s mean, 12
calls), clef 5/5 (11.5s, 12 calls), luna unchanged at 5/5 (14.9s).
* bench(testbed): Laya through systemOne
Laya serves the System One API and accepts structured values, so it runs
through systemOne({ structured: true }) unchanged. On the todo flow its
first operation question chose unsupported at 0.29 / 0.10 on four runs,
answered by the multilingual checkpoint its router picked for an English
prompt; the fifth run and further probes hit insufficient_credits.
* fix(decision): address review on rounding, history, and evaluation ids
- validateDecision accepts a rounded distribution only when 1 lies between
the sums of each value's rounding interval, clipped to [0, 1]; a summed
per-option tolerance exceeded 1 with many options and admitted two
options at probability 1.
- Decisions send the 20 most recent history entries, each clipped to 240
characters, so long steps and large params stay within request limits.
- evaluationModel() builds its answer, confidence, and criteria maps with
null prototypes, so an id such as __proto__ stays an own entry.
- @ai-sdk/provider moves to dependencies: the public declarations import
its evaluation types. ai stays an optional peer.
* bench(testbed): pin Laya's English checkpoint
Laya's router sent the full English prompt to its multilingual checkpoint.
The API accepts model 'english' to pin the English one (its routing
metadata reports the explicit model), so the bench config sends it.
* docs(decision): benchmark Laya locally on M1, both checkpoints 0/5
* docs(decision): reconcile laya narrative with run evidence, money not tokens
* feat(decision): caller-overridable transport flags on clef() and jev()
* feat(decision): AI SDK evaluate executor with element table and text model
* fix(decision): empty-tree check and provider pin
* fix(decision): address review on secrets, options, errors, and docs
Answers the cubic threads on the evaluate-only executor: terminal checks reject an incomplete screen instead of judging it, secret markers are stripped at any depth, hidden/disabled options are never offered, select options share the 255-choice cap, select criteria name the option, out-of-range probabilities fail, text-model errors mirror the decision mapping, and the README/docs describe the evaluate-only API.
* fix(decision): count dropped select options and fix README
Counts enabled options inside rows dropped past the element cap as omitted, adds the E2EConfig type import to the README example, and qualifies when type is offered.
* fix(decision): settle terminal claims by the check and match codebase style
The second rejected done/failed claim now ends by the check verdict (fails -> ACTION_FAILED, anything else -> blocked), and the completion check honors the gates. The decision state carries the last 10 actions, check targets show their checked state, checked radios are never toggled, and options dropped by the per-question cap stay out of omitted. The docs example opens /todos in test code since navigate is not offered. Single quotes, template literals, and split long lines across the package.
* fix(decision): show filled fields to verdicts and pin the history window
Verdict views (assert and the completion check) build the element table with typing enabled, so a field the step filled keeps its row and value even when the step itself cannot type. The history test now asserts the window holds the newest ten actions.
* fix(decision): let completion checks see results and name the text key
Page text keeps a named node's text when it differs from the name, as formatNode does, so a status named Greeting that says Welcome back, admin! reaches the verdict. The completion check now asks whether the task is complete and sees the non-secret params and the actions with their typed values; rejected claims stay out. The text prompt names the single text key, asks for no code fence, and prefixes the request so JSON-object response modes accept it.
* fix(decision): keep secret handles out of replayed history and mark check inputs as data
Replayed summaries name a filled secret handle; seeding the history now replaces declared secret names with <secret>, so names reach the model only as secret question criteria. The completion check treats action descriptions, typed values, and params as data, and the guide says the check sees only non-secret values.
* fix(decision): redact only the handle in replayed secret-fill summaries
Matching the declared secret name anywhere in a replayed summary also erased labels and values equal to it, and missed handles the runner cut at 40 characters. Only the quoted handle after fill secret at the start of the summary becomes <secret> now.
* docs(decision): describe the one-request fan-out in the guide intro
* fix(decision): count a scroll that changes what is in view as progress
The fingerprint behind the stall guard hashed node content but not position, so a scroll that only moved the viewport read as page unchanged and the third one blocked the step. It now also hashes whether each node meets the viewport (membership, not coordinates), so a scroll that brings other nodes into view resets the count while one that moves nothing, as at the page bottom, still trips the guard.
* chore: prune @pnpm/exe from the lockfile after merging main
* fix(decision): match the supported Node range in engines
---------
Co-authored-by: Oskar Kwaśniewski <oskar@okwasniewski.com>
* fix(e2e): load TypeScript with an oxc loader instead of tsx
* fix(e2e): close loader edge cases: sourceless CommonJS, export maps without extensions, unreadable tsconfig, chained source maps, Windows require paths
* fix(e2e): refuse Bun and Deno runtimes, name oxc load failures, follow Yarn PnP export targets
* perf(e2e): skip the syntax check for files whose only @ is in a package import
* fix(e2e): allow type-only exports in .cts, warn from runner only, Bun check first, docs
* fix(e2e): workers warn about tsconfigs runner did not, FILE: urls, foreign infra errors, review tests
* fix(skill): declare the app on the target in skill and docs config examples
The bundled skill and docs/authentication.mdx still taught web({ url, command }), removed in #698. The skill's two config examples now live in docs/examples/skill/, typechecked and mirrored by check-docs-examples, so the old shape fails pnpm check.
* fix(skill): run the docs drift check on skill changes, drop stale service wording
* feat!: browser fixture, getByRole name arg, toMatchSchema, top-level hooks, separate credential namespace
API review feedback before launch:
- web fixture renamed browser (Web -> Browser, WebExpectation ->
BrowserExpectation); requires: ['browser']; destructuring web hints the
new name. web() factory and other Web* types unchanged.
- getByRole(role, name, options?) beside the object form; MCP locate
writes the short form.
- expect(value).toMatchSchema(standardSchema): sync, returns the typed
output; poll resolves to it, soft returns undefined after a kept failure.
- describe, beforeEach, afterEach, beforeAll, afterAll exported from e2e;
web and mobile export typed beforeEach/afterEach.
- credentials and secrets are separate namespaces: a password handle is
<name>.password, secrets.get() never returns it. Committed recordings
migrated offline (old key hash verified, then recomputed).
- runtime test.skip keeps Playwright semantics; docs say it throws.
* docs(web): browser acts on one active tab; cookies, routes, dialogs cover the whole browser
* fix(secrets): judge a fill's purpose by the config, not the handle; no old-name hints
A config-time secrets.get() handle reads its purpose from the run's
registry again, and authorizeSecretFill checks the registered purpose, so a
handle claiming generic-secret cannot put a password into a plain field.
Drop the web fixture migration hint. Changesets name the on-screen marker
change for replay entries.
* docs(skill): browser fixture, getByRole name arg, toMatchSchema, top-level hooks, credential namespace on the rewritten skill
* fix(expect): negated toMatchSchema returns nothing, as its type says
* fix(expect,docs): soft double negation stays soft, null ~standard is INVALID_ARGUMENT, Playwright samples keep Playwright's getByRole
Also documents SoftExpectCall, SoftValueExpectation, and NegatedPollExpectation
on the expect reference, requires: ['browser'] in the skill's options table,
and tightens the poll and secret-purpose tests.
* docs(babysit): triage pattern for scripted renames that rewrite another library's code
* docs: engine options take a secrets entry, web package hook exports, vault example imports
* refactor(secrets)!: credentials hold their password, allSecrets feeds fills; engines export describe and every hook; BrowserRoute, BrowserResponse
From the maintainability review:
- ResolvedConfig.secrets is the config.secrets entries only; a credential
carries its password; allSecrets is the one map fills, the ledger, and
redaction read. The per-consumer purpose filters and the .password
suffix parsing are gone, and e2e mcp builds handles from config.
- The config load and a running test name the credential a missing
secrets.get() most likely meant.
- @e2e-dev/web and @e2e-dev/mobile export describe and all four hooks
beside test, so a test file registers from one import.
- WebRoute and WebResponse are BrowserRoute and BrowserResponse.
- getByRole normalizes its arguments once; isTextMatch is the one
string-or-RegExp check; poll reuses ValueMatcherName.
* revert(web): keep WebRoute and WebResponse; Browser names the fixture, Web the network and option types
* feat(config)!: remove specVersion from the config
The runner version is the format version. A config that still sets
specVersion fails INVALID_CONFIG telling the author to remove it. The
report's run.specVersion and the replay cache key keep their hard-coded
copies.
* feat(collect): "!" exclusions in the tests globs
An entry starting with "!" removes files an including glob matched,
in any order. Discovery never reads a directory an exclusion takes
whole, unless an including glob could reach a dot name beneath it.
A list of only exclusions is INVALID_CONFIG.
* refactor!: CacheStore and web({ screencast }) renames
TraceCacheStore is CacheStore, and the docs, skill, and CLI call the
cache the replay cache. web({ video: { size, quality } }) is
web({ screencast }) with WebScreencastOptions; web({ video }) fails
INVALID_CONFIG naming screencast.
* feat(secrets): secrets.get() in engine options, resolved per attempt
At config evaluation, before any run exists, secrets.get(name) returns
a reference by name. An engine declares the handles its options hold
in defineEngine({ secrets }); the config load fails INVALID_CONFIG for
a name no secret or credential has. The engine reads the value through
EngineAttemptContext.resolveSecret, which runs a provider fresh and
registers the value in the session and process ledgers before it
returns, without tainting the viewport. e2e/engine exports isSecret.
web({ basicAuth: { password } }) takes a Secret and declares it. Docs
show process.loadEnvFile('.env') in the config.
* feat(run)!: trace and video share recording modes; artifacts holds only store
One RecordingMode ('off' | 'on' | 'retain-on-failure' | 'on-first-retry' |
'on-all-retries') for trace and video, set on the config, a target, the CLI
(--trace [mode], --video [mode]) and a test, one precedence rule: config,
target, CLI, test. Trace defaults to on locally, on-first-retry in CI, with
a plan-time notice when a retry mode meets retries: 0. Serial members cannot
set trace.
Capability rule: a mode from the config root or a flag applies to targets
whose engine can record it, one notice naming the rest; a mode set on a
target or a test is required and fails with UNSUPPORTED_ARTIFACT.
artifacts.kinds (both forms) and artifacts.trace are removed and fail
INVALID_CONFIG naming the trace mode they meant. Failure screenshot and
screen text are always captured when the engine can. MCP sessions keep the
target's trace mode and trace only under 'on'.
* feat(run)!: output directory for every run result; ArtifactStore.putLink
output (default .e2e) and --output <dir> on run and explore hold report.json,
junit.xml, summary.md, ai-trace.json, artifacts/, sessions/ and the e2e mcp
session files; --last-failed reads <output>/report.json. The reportSibling
derivation from the artifacts dir is gone. --artifacts stays as a hidden
flag that fails with the --output it maps to.
A run clears <output>/artifacts at start. cache.dir stays independent. An
output that is the project root or outside it, is or sits in cache.dir,
would hold cache.dir under artifacts/, sessions/ or videos/, or holds a
tests glob root fails INVALID_CONFIG with the reason.
ArtifactStore gains optional putLink(link) for provider-hosted video links;
its ref lands beside the url. A passed retain-on-failure attempt's link is
never handed over, and a failing putLink never fails the run.
bench-ab and the mobile workflow move to --output.
* test(secrets): a hold's registry is probed by an unknown name, now that get() defers without one
* test(web): the download trigger's measured wait tolerates a timer firing early
* fix(secrets): rewrite the trace of an attempt whose engine holds a secret
A Playwright trace records the context options, basic-auth password in
plaintext, but a trace was rewritten only after a fill. resolveSecret now
marks the session so its trace goes through the redactor, screencast
frames kept since the value never reached the screen. The fake engine
writes a real trace archive, and the basic-auth fixture escapes its echo
and refuses a malformed header.
* fix(web): race the basic-auth secret with the attempt's signal
A provider still resolving after the launch timeout set the session after
the runner ended the attempt, so the retry failed with an attempt already
running. Resolution now runs first, raced with the signal.
* fix(mcp): a session notes a configured trace its engine cannot record
* test: a shadow-free fake trace archive and an ungarbled exclusion title
* docs: output paths, replay cache wording, and review corrections
--last-failed, --ai-trace, and the json reporter name <output> paths;
the skill says output never moves cache/ or the app's log; the
INVALID_GLOB row states the check the code makes; serial members list
only; the engine vocabulary names Secret and isSecret; the basic-auth
example uses a provider; the last trace cache wording is replay cache.
* chore(changeset): leave released notes alone; pending notes match the renames
Released CHANGELOG entries keep TraceCacheStore, the name they shipped
with; cache-store-rename.md announces the rename. The pending video and
provider recording changesets name on-all-retries and web({ screencast }).
* chore(bench): refuse a forwarded --output the harness would not clean up
* fix(secrets): a rewritten trace drops its screencast frames, engine-held secrets included