gflow-cli 0.53.0__tar.gz → 0.54.0__tar.gz

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "gflow",
-  "version": "0.53.0",
+  "version": "0.54.0",
   "description": "Reusable development and operations workflows for gflow-cli",
   "author": {
     "name": "Flavio Oliva",

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/.github/dependabot.yml

@@ -51,14 +51,42 @@ updates:
     # `groups` unless a group sets `applies-to: security-updates`, so an advisory
     # fix still opens immediately on its own instead of waiting for Monday.
     ignore:
-      # playwright minors are gated by the `<1.60.0` upper bound in
-      # pyproject.toml, not by this rule — an untested driver minor silently
-      # wedged video generation (see the bound's comment). This entry keeps
-      # majors out even if that bound is ever widened. Raising playwright is a
-      # deliberate, live-verified act; the weekly `deps-watch` workflow reports
-      # how far behind the pin has drifted so it does not freeze unnoticed.
+      # playwright MINORS must be ignored here. They are NOT gated by the
+      # `<1.62.0` upper bound in pyproject.toml -- that was the assumption this
+      # rule used to state, and PR #465 falsified it. The "uv" ecosystem does
+      # not respect a constraint standing in an update's way, it REWRITES it:
+      # #465 widened `playwright>=1.61.0,<1.62.0` to `<1.63.0` and locked
+      # 1.62.0 -- the exact version observed hanging every `video i2v` right
+      # after the frame upload (see the bound's comment in pyproject.toml).
+      # Only tests/test_playwright_pin.py stopped it, and on its way out it
+      # red-lighted the whole grouped weekly batch.
+      #
+      # PATCH bumps stay allowed on purpose: the 1.61.x headroom exists so a
+      # driver CVE fix does not require a gflow release. Raising a MINOR is a
+      # deliberate, live-verified act (AGENTS.md "verify before you act") and
+      # also means hand-editing SUPPORTED_PLAYWRIGHT_RANGE in
+      # src/gflow_cli/api/transports/ui_automation_video.py, which no bot can
+      # do -- so a bot-authored minor bump can never be mergeable here anyway.
+      # The weekly `deps-watch` workflow reports how far behind the pin has
+      # drifted so it does not freeze unnoticed.
       - dependency-name: "playwright"
-        update-types: ["version-update:semver-major"]
+        update-types:
+          - "version-update:semver-major"
+          - "version-update:semver-minor"
+      # patchright is the SAME hazard with a bigger blast radius, and the exact
+      # `==` pin protects it no better than playwright's upper bound protected
+      # playwright -- an exact pin is just another constraint for the "uv"
+      # ecosystem to rewrite. It ships a PATCHED Chromium driver that loads the
+      # real Google session (cookies, OAuth Bearer, SAPISID), so docs/SECURITY.md
+      # requires every bump to get a security review and explicitly says it is
+      # "NOT a routine dependabot auto-merge". That was prose with nothing
+      # enforcing it; this is the enforcement. ALL update-types, patch included:
+      # any version change moves the browser binary underneath the user.
+      - dependency-name: "patchright"
+        update-types:
+          - "version-update:semver-major"
+          - "version-update:semver-minor"
+          - "version-update:semver-patch"
 
   # GitHub Actions
   - package-ecosystem: "github-actions"

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/.github/workflows/deps-watch.yml

@@ -66,7 +66,7 @@ jobs:
         # Unlike the step above, this ignores our constraints and reports the
         # true latest on PyPI — so a package frozen by one of OUR upper bounds
         # shows up here and nowhere else. This is the expiry alarm for those
-        # bounds: `playwright>=1.59.0,<1.60.0` is deliberate (an untested driver
+        # bounds: `playwright>=1.61.0,<1.62.0` is deliberate (an untested driver
         # silently wedged video generation — see the comment in pyproject.toml),
         # but a deliberate pin still needs a periodic "how stale is it now?"
         # signal, and raising it requires a live-verified generation first.

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/.github/workflows/external-pr-triage.yml

@@ -15,7 +15,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Triage external pull request
-        uses: actions/github-script@v7
+        uses: actions/github-script@v9
         with:
           script: |
             const pr = context.payload.pull_request;

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/.github/workflows/labels.yml

@@ -37,7 +37,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Create or update labels
-        uses: actions/github-script@v7
+        uses: actions/github-script@v9
         with:
           script: |
             // Every label referenced by .github/dependabot.yml. Keep in sync.

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/CHANGELOG.md

@@ -7,6 +7,27 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.54.0] — 2026-08-12
+
+### Changed
+
+- **Clearer close-the-browser guidance in `gflow auth login` (#470).** Reworded the final passive-capture step in `real_chrome.py` from the abrupt "CLOSE THE BROWSER" command into plain-language guidance: closing the Chrome window is how you signal you're done, after which gflow verifies the Flow session automatically. No behavior change — the manual close stays required on the real-Chrome path, because Chrome holds an exclusive lock on its cookie store while running (verified empirically), so gflow cannot auto-detect completion there without breaking the zero-automation-surface stealth model.
+- **Overlay/watermark detection hardened to pure structural selectors.** Council-review cleanup of `ui_automation.py`: enforced 100% language-agnostic structural anchors, removed text-label hacks and dead aliases, and fixed a stale test assertion — strengthens the #403 release-modal dismissal across localized profiles.
+
+### Fixed
+
+- **Dependabot could re-offer the known-bad playwright 1.62.0 every Monday (#465).** The `uv` ecosystem does not respect a version bound standing in an update's way — it **rewrites** it, so the deliberate `playwright>=1.61.0,<1.62.0` bound in `pyproject.toml` never gated Dependabot as `.github/dependabot.yml` claimed. PR #465 widened it to `<1.63.0` and locked 1.62.0, the exact version documented as hanging every `video i2v` right after the frame upload. Dependabot now ignores playwright `semver-minor` alongside `semver-major` (patch bumps stay allowed so a driver CVE fix needs no gflow release), guarded by `tests/test_playwright_pin.py::test_dependabot_ignores_playwright_minor_bumps`.
+- **`patchright` was exposed to the same bug, with a bigger blast radius.** Its exact `==` pin is just another constraint the `uv` ecosystem can rewrite, so `docs/SECURITY.md`'s "a Patchright version change is NOT a routine dependabot auto-merge" was prose with nothing enforcing it. Dependabot now ignores **all** patchright update-types (any version change moves the patched Chromium driver — which loads the real Google session — underneath the user), guarded by `test_dependabot_ignores_driver_engine_bumps`. Also corrected the stale pin cited in `docs/SECURITY.md` (`1.60.1` → `1.61.2`) and the stale playwright range in `.github/workflows/deps-watch.yml`.
+- **Character-composer focus isolation.** `_submit_body_prompt` now preserves the locator-scoped `press_sequentially`, keeping correct focus during character-composer submission.
+
+### Security
+
+- **Timing-jitter entropy now uses a cryptographically-strong RNG (SonarCloud S2245).** The `_jitter_ms` interaction-delay humanization helper draws from `secrets.SystemRandom` instead of the standard `random` module, so the jitter entropy is not predictable.
+
+## [0.53.1] — 2026-08-06
+
+- Re-release of 0.53.0 (version bump only; recorded headed live-verification evidence). No functional changes.
+
 ## [0.53.0] — 2026-08-06
 
 ### Added
@@ -2753,7 +2774,9 @@ shell-script template that branches on these codes.
 
 First skeleton. Not functional end-to-end yet.
 
-[Unreleased]: https://github.com/ffroliva/gflow-cli/compare/v0.53.0...HEAD
+[Unreleased]: https://github.com/ffroliva/gflow-cli/compare/v0.54.0...HEAD
+[0.54.0]: https://github.com/ffroliva/gflow-cli/compare/v0.53.1...v0.54.0
+[0.53.1]: https://github.com/ffroliva/gflow-cli/compare/v0.53.0...v0.53.1
 [0.53.0]: https://github.com/ffroliva/gflow-cli/compare/v0.52.0...v0.53.0
 [0.52.0]: https://github.com/ffroliva/gflow-cli/compare/v0.51.0...v0.52.0
 [0.51.0]: https://github.com/ffroliva/gflow-cli/compare/v0.50.0...v0.51.0

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: gflow-cli
-Version: 0.53.0
+Version: 0.54.0
 Summary: Unofficial CLI for Google Flow — drive Veo image-to-video generations from the terminal.
 Project-URL: Homepage, https://github.com/ffroliva/gflow-cli
 Project-URL: Issues, https://github.com/ffroliva/gflow-cli/issues
@@ -58,7 +58,7 @@ Requires-Dist: numpy>=1.24; extra == 'dev'
 Requires-Dist: pyright>=1.1.0; extra == 'dev'
 Requires-Dist: pytest-asyncio>=0.23.0; extra == 'dev'
 Requires-Dist: pytest>=8.0.0; extra == 'dev'
-Requires-Dist: ruff==0.16.1; extra == 'dev'
+Requires-Dist: ruff==0.16.2; extra == 'dev'
 Provides-Extra: gcs
 Requires-Dist: gcsfs>=2024.2.0; extra == 'gcs'
 Requires-Dist: universal-pathlib>=0.2.5; extra == 'gcs'

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/docs/INDEX.md

@@ -135,7 +135,7 @@ Slash commands for Claude Code, stored in `.claude/commands/gflow/`. All prefixe
 **"A command failed — where is the incident bundle and what's in it?"** → [DEBUGGING § Automatic incident bundles](DEBUGGING.md#automatic-incident-bundles) (layout, triggers, review-before-sharing); privacy boundaries: [SECURITY § Automatic incident bundles](SECURITY.md#automatic-incident-bundles-gflow_cli_incident_capture-default-on); disable via [CONFIGURATION § GFLOW_CLI_INCIDENT_CAPTURE](CONFIGURATION.md#gflow_cli_incident_capture)
 **"Flow's UI broke a selector — how do I diagnose it?"** → [DEBUGGING § Inspecting Flow's live UI](DEBUGGING.md#inspecting-flows-live-ui)
 **"What does each `ui_automation.*` log event mean?"** → [DEBUGGING § Listener & HTTP-layer debugging](DEBUGGING.md#listener--http-layer-debugging)
-**What was actually live-verified for the latest release?** → latest: [LIVE_VERIFICATION_v0.52.0](LIVE_VERIFICATION_v0.52.0.md) (**batch DAG references #317, character entity provenance recording #402, and video duration selector cascade fix #451 live-verified 2026-08-05** — 3/3 live E2E tests in `tests/e2e/test_entity_provenance_e2e.py` passed clean on profile `ffroliva`, 5-layer evidence ledger verified). Prior: [v0.51.0](LIVE_VERIFICATION_v0.51.0.md) (**dependency/security/test-correctness release verified 2026-08-05, 1 Imagen credit + 1 Veo attempt** — the only change needing live evidence was the `playwright` bound raise `>=1.59.0,<1.60.0` → `>=1.61.0,<1.62.0`, and the 2026-08-03 regression does NOT reproduce on 1.61.0: a live i2v drove `image_uploaded status=200` → `frame_attached` → `generate_captured status=200` with `startImage` parsed, where **1.62.0 hung silently at exactly that upload step**; live i2i local-ref attach passed outright [96.7s]. **1.62.0 stays excluded**, never root-caused. `-m e2e_auth` 16/16, daemon e2e 1 passed [MCP over Streamable HTTP — the first live coverage that test has ever had, it was structurally incapable of passing before], MCP Tasks 5/5 layers, 2923 offline tests, `pip-audit --all-extras` clean. **No finished mp4 is claimed** — Flow's own generation returned `PUBLIC_ERROR_VIDEO_GENERATION_TIMED_OUT`, server-side. Defect found and recorded not omitted: Flow dropped the duration-tab UI so **`--duration` is currently broken for i2v**, A/B-proven pre-existing on BOTH 1.59.0 and 1.61.0 — [#451](https://github.com/ffroliva/gflow-cli/issues/451)). Prior: [v0.50.0](LIVE_VERIFICATION_v0.50.0.md) (**MCP Tasks extension SEP-2663, non-blocking task handle responses, CLI/MCP output path hardening -o/--output, and Telegram PR triage alert resilience verified 2026-08-04** — 5-layer evidence ledger note generated at `tmp/live-verify/mcp-tasks.md` confirming instant task handle return, working status queries, and cancelled status DB state transitions; full 118-test suite green). Prior: [v0.49.0](LIVE_VERIFICATION_v0.49.0.md) (**#125 `omni-flash` start-frame I2V video generation, Playwright dependency upper-bound pin `>=1.59.0,<1.60.0`, submission stage watchdog timeout, and #404 count-tab fail-closed logic verified 2026-08-03** — 0-credit wire intercept captured `batchAsyncGenerateVideoStartImage` with `startImage` bound; live costed generation on `omni-flash` I2V produced a 1.2MB valid mp4 video asset on `ffroliva`; full 2902-test release suite green). Prior: [v0.48.0](LIVE_VERIFICATION_v0.48.0.md) (**#411 predictable `-o`/`--output` paths, offline-verified 2026-08-02, ZERO credits** — exact single-target local writes + `_1`/`_2` multi-count **image** suffixes proven by the CLI suite; the pre-tag council audit [RED, then resolved] re-scoped the release to the truthful surface: a dead MCP `output` param was deleted [worker queue never read it — #414], cloud `-o` targets and video multi-count suffixes are follow-ups [#415], and two tautological video-path tests are recorded as weak evidence rather than claimed as proof; storage-layer S3 writes verified against Docker MinIO at the layer level only; 2888-test release tree green; the live-Flow run aborted pre-generation on an expired denon82 session [`AuthExpiredError` 401, correct RFC-7807 contract + remediation hint, $0] — recorded, not omitted; live re-run queued for the next authenticated session). Prior: [v0.47.0](LIVE_VERIFICATION_v0.47.0.md) (**MCP SDK 2.0.0 migration + dual-era protocol + #402 entity provenance recording live-verified 2026-08-01, ZERO credits** — stdio JSON-RPC initialization handshake verified over subprocess transport, `resolve-drift` CI job added to prevent dependency drift, and 121 unit tests in `tests/data` passed). Prior: [v0.46.1](LIVE_VERIFICATION_v0.46.1.md) (**#404 classic count-setter fix live-verified 2026-07-31 on the reporter's exact environment, ZERO credits** — credit-free DOM recon proved Flow renamed the count-tab labels to uniform `xN` [`x1` replaces `1x`, image AND video rows], that `nth(0)` of the old filtered set landed on the selected `x2` and actively moved the count AWAY from the target, and that a digit-keyed click on `x1` converges instantly and persists without a Save commit; the new classic e2e then passed live [count=1 against a fresh project displaying 2 → exactly 1 image; count=2 early-exit → exactly 2], and the #313 agentic count-enforcement e2e passed 4/4 [counts 1–4 vs deliberately mismatched sticky defaults, magic-byte-verified downloads] after its stale setup was repaired [Flow now loads the editor in classic mode — the setup needed an Agent-pill toggle before the `tune` icon; test-only fix, no product code implicated]; the video output-count cascade carries a recorded waiver [full drive costs Veo credits; its exact selector shape was exercised live in the spike's cleanup step]). Prior: [v0.46.0](LIVE_VERIFICATION_v0.46.0.md) (**#387 provider-agnostic prompt-tools transport live-verified 2026-07-28, ZERO credits** — a four-probe capability spike passed 4/4 against BOTH a non-Google gateway and Google's OpenAI-compat endpoint, including a *semantic* multimodal check [a generated solid-red JPEG round-tripped and the model answered 'Red'], which is what proved the native Gemini `inlineData` path could be deleted outright rather than kept as a fallback; the real CLI then returned `was_expanded: true` on gateway/gateway-no-model/Google-default, emitted the loud one-time notice + no-op for the removed `GFLOW_CLI_GEMINI_API_KEY`, and made NO network call when unconfigured. Live verification caught two defects a fully green 2849-test offline suite missed: Google's compat endpoint has no server-side default model [`400 "model is not specified"`, silently no-oping the DEFAULT configuration], and the first redirect guard did not actually work because `build_opener()` installs its default handlers regardless of arguments, so `Authorization` would still have been replayed cross-host. `/code-review high` is recorded as NOT run — it is user-triggered only). Prior: [v0.45.0](LIVE_VERIFICATION_v0.45.0.md) (**#393 UUID `--ref` binding + #395 character-entity binding live-verified 2026-07-27/28** — a controlled A/B on identical input flipped `--ref` from exit 9 to exit 0 via the catalog-backed upload fallback, with the anti-silence guard [an unbindable ref aborts non-zero and writes no image] pinned by a live e2e; picker DOM + HAR showed Flow indexes a short auto-caption, NOT the prompt, so a prompt-derived search hint was dropped rather than shipped on assumption; a CDP spike against Flow's own character flow then captured the `entityContext` contract that gflow's generations were missing, root-causing #395 to two gflow defects [overlay dismissal Escaping Flow's own composer, and a character-route bounce that sent the prompt to the project composer] — both fixed, both character e2e tests now pass live, including `--format-prompt` [#383], with a read-back confirming the character carries its name and `thumbnail_media_id`). Prior: [v0.44.0](LIVE_VERIFICATION_v0.44.0.md) (**#381 dual-side project management & Playwright Chromium transport live-verified 2026-07-26** — real-editor project creation + rename tRPC server/DOM sync, prompt slugging, `gflow project` subcommand family, HTTP 429 adaptive backoff, and full E2E test suite). Prior: [v0.43.0](LIVE_VERIFICATION_v0.43.0.md) (**private incident diagnostics live-verified 2026-07-23, 1 Veo credit**). Prior: [v0.42.0](LIVE_VERIFICATION_v0.42.0.md) (**#361 production-readiness live-gaps confirmed live 2026-07-21, 1 Veo credit** — a real `veo-lite` t2v through the worker path persisted a real `remote_started` checkpoint [observer→checkpoint→DB chain confirmed], crash-recovery classified the task `indeterminate` without resubmitting, and mid-launch cancellation released the `ProfileLease` credit-free; full evidence in [v0.40.0-production-readiness § "2026-07-21 live follow-up"](LIVE_VERIFICATION_v0.40.0-production-readiness.md#2026-07-21-live-follow-up-0420-cycle); #359 content-safety `400`→`ContentPolicyError` classification is unit-covered, not live-triggered [would require a deliberate policy violation]; #360/#362 are docs-only, verified by the documentation gates). Prior: [v0.41.0](LIVE_VERIFICATION_v0.41.0.md) (**#357 production-readiness hardening — stale-session fail-fast, free image gen, paid veo-lite T2V all proven live**). Prior: [v0.40.0](LIVE_VERIFICATION_v0.40.0.md) (**#344 `@`-mention asset-tagging resolution proven live e2e, ~2 Imagen credits** — `gflow character create` staged a real taggable character [bound reference images, `has_reference_images` guard], the e2e test's documented propagation wait cleared Flow's `projectInitialData` lag, and `gflow image t2i "@Zoro …"` passed with returncode 0, a real generated image, and the correctly de-tagged prompt in the catalog — no code changes needed, `resolve_and_apply` as merged in #344 already works end-to-end; pre-existing unrelated issue #174 [Flow media-library A/B flake] can still intermittently block generation regardless of this feature — recorded, not omitted). Prior: [v0.39.0](LIVE_VERIFICATION_v0.39.0.md) (**#341 failed-generation persistence proven live** — a real `image t2i` on denon82 failed at the wire [`batchGenerateImages` HTTP 400 → `WireFormatError`] and the record-on-catch funnel wrote the **first-ever `failed` row** into the production catalog [`error_type=wire-format`, redacted `error_detail`, browsable via `gflow data list errors`], **$0**; the remaining paid paths [video/chain/batch/movie/worker] are funnel-regression-tested, not force-failed live [would cost credits], and the status-aware `succeeded=false` video-poll fix is unit-locked — recorded, not omitted). Prior: [v0.38.1](LIVE_VERIFICATION_v0.38.1.md) (**#338 agentic-pin recovery proven live ON the pinned account** — run 1 caught the composer-render race the unit suite could not [`ensure_media_incomplete` +115 ms, toggle never reached]; the readiness-wait fix landed and run 2 recovered classic on denon82 after ~2h of pin [`ui_driver.bound mode=classic`, real 768×1376 JPEG, exact prompt adherence]; reload branch unit-locked, live trigger not observed this cycle — recorded, not omitted). Prior: [v0.38.0](LIVE_VERIFICATION_v0.38.0.md) (**#332 mode control + #314 i2i ref dedup proven live e2e** — `image t2i --ui-mode classic` recovered agentic→classic in a real run [`ui_mode.attempt_exit_agent` → full classic chain, 1376×768 JPEG]; two identical `image i2i` runs proved the dedup contract [run 1: picker miss → `image_uploaded`; run 2: `reference_deduped_by_filename`, ZERO uploads]; `video t2v` veo-fast [`MEDIA_GENERATION_STATUS_SUCCESSFUL`, 15.2 MB `ftyp` mp4]; autopilot #333 recorded deferred-with-reason [hermes-ops deployment staged separately]). Prior: [v0.37.0](LIVE_VERIFICATION_v0.37.0.md) (**viewport 1920×1080 + agentic count enforcement + FIPS-safe SAPISIDHASH proven live e2e** — `image t2i` [classic cohort, real 1376×768 JPEG] and `video t2v` veo-fast [`MEDIA_GENERATION_STATUS_SUCCESSFUL`, 11.5 MB `ftyp` mp4] both succeeded end-to-end at the new 1920×1080 viewport with every selector resolving [no drift]; SAPISIDHASH digest verified byte-identical [FIPS-safe `usedforsecurity=False`], auth unchanged). Prior: [v0.36.0](LIVE_VERIFICATION_v0.36.0.md) (**#316 diagnostic tooling live-verified against real Flow traffic** — `GFLOW_CLI_HAR_PATH` captured a real 54-entry HAR [real Flow domains, 34× 200s, and a genuine 401 the tool exists to diagnose]; `GFLOW_CLI_DEBUG_TRACEBACK`'s typed-error scoping confirmed live [console output identical with/without the flag on a `GFlowError`]; unhandled-path behavior covered by 6 new tests (30 total across both files) + SDD task review; #312 reference-entity-smuggling fix covered by its own dedicated e2e test). Prior: [v0.35.0](LIVE_VERIFICATION_v0.35.0.md) (**#305-follow-up multimodal reverse-engineering & storyboard tool proven live e2e** — `gflow tools run storyboard` successfully deconstructed a cyberpunk scene into a cohesive 4-panel sequence; `reverse-engineer` tool extracts frames from video streams/URLs using `claude-video`'s `watch.py` and produces structured prompt components using Gemini multimodal models; dynamic token budgeting and strict Pyright types verified via test suite). Prior: [v0.34.0](LIVE_VERIFICATION_v0.34.0.md) (**#299 `--ui-mode` gate proven live, ZERO credits** — drove `get_ui_driver` directly over 4 real editor loads, never generated: bidirectional switch works [classic→agentic via `_force_agent_mode`, agentic→classic via `_exit_agent_mode`, each DOM-reverified], `agentic`-when-already-agentic is idempotent, and a timed-out toggle → **exit-28 fail-fast before submission**; deprecation behavior-change [`prefer_classic`→exit 28] confirmed). Prior: [v0.33.0](LIVE_VERIFICATION_v0.33.0.md) (**#287-follow-up picker fix proven live on the original repro** — the v0.32.1-failing `i2v <uuid> --project <crowded>` command now exits 0 with a real 3.0 MB `ftyp` mp4, `frame_ref_attached` + `video_saved`, zero uploads; **#241 jitter live-verified on both paths** — flag `--jitter 4-6` → `batch_jitter_sleep seconds=4.98`, and the new 0.5–1.5 s default → `seconds=1.47`, each firing exactly once between two t2i submissions; environment note: the denon82 cohort is now natively agentic and round-2 generations hit the pre-existing #281 `MediaAttributionError`, not a regression). Prior: [v0.32.1](LIVE_VERIFICATION_v0.32.1.md) (**#293 teardown fix live-verified credit-free** — error-path aborts leave ZERO chrome.exe holding the profile [3× leak reproduced pre-fix same day]; force-close fallback + ProfileLockedError translation unit-pinned; #283 wave logic fixes mock-verified to exact call counts). Prior: [v0.32.0](LIVE_VERIFICATION_v0.32.0.md) (**#287 UUID i2v frames proven live e2e** — an in-project asset selected by media UUID in the frame-slot dialog [frame_ref_attached, startImage bound on the wire, ZERO upload events] produced a real 720×1280 interpolation, resolving the #237 negative prior; foreign UUID → exit 9 pre-generation with screenshot; **#288 duration fail-fast proven live** — exit 23 pre-submit saved a 10-credit generation, and the debug screenshot confirmed the duration control is ABSENT from the cohort's settings popover [locale hypothesis refuted]; upload-400 exit-27 repro deferred-with-reason — the original rejected JPEG no longer exists). Prior: [v0.31.0](LIVE_VERIFICATION_v0.31.0.md) (**#281 wrong-media defenses live-verified in BOTH directions, credit-free** — correct attribution downloads with real wire metadata [seed, 768×1376]; the incident-class lazy-render ambiguity reproduced live and hard-stopped with `MediaAttributionError` exit 26 naming every candidate UUID, no file written; **#282 two-UUID-ref i2i proven e2e** — both freshly generated portrait UUIDs picker-selected into one scene with both canonical identities visually transferred). Prior: [v0.30.0](LIVE_VERIFICATION_v0.30.0.md) (**agentic-cohort image path + both fixes live-verified credit-free** — native 768×1376 stills, i2i-chain + canonical-ref identity hold; the character-create `DataIntegrityError` and the `prefer_classic` WARNING were observed live then fixed [recorder idempotent + INFO log]; MCP video-param feature deferred-with-reason on Veo credits, per the stills-only goal). Prior: [v0.29.0](LIVE_VERIFICATION_v0.29.0.md) (**persistent `gflow instructions` CRUD proven live e2e GREEN, credit-free** — 11-command sequence on a fresh project: add incl. `--ref <image-UUID>` attach, list/enable/disable, master toggle both ways, TOML `apply` destructive full-sync with `enabled=false` honored, `rm` with server card-id preserved across mutations; movie brief-sync code-path-verified via the same primitives, deferred-with-reason on credits; MCP `gflow_instructions_*` tools are thin adapters over the live-verified primitives). Prior: [v0.28.0](LIVE_VERIFICATION_v0.28.0.md) (**agent instructions `-i` now steer agentic generation — crayon e2e GREEN**: a style-neutral prompt + an enabled "crayon" card produced an unmistakable crayon drawing through the real transport; root causes were conversational phrasing + the `project_brief.enabled` master switch). Prior: [v0.27.1](LIVE_VERIFICATION_v0.27.1.md) (**v0.27.0 release follow-up fixes and documentation sync**). Prior: [v0.27.0](LIVE_VERIFICATION_v0.27.0.md) (**`[style]` block with named variants + prompt-aware resume — credit-free CLI verification**: real `movie run --dry-run` 3-way ledger proving resolved-style display, `skip (done)` on matching `style_hash`, and `re-run (style changed)` after a style edit; composition is local, wire path untouched, so no Veo credits spent — first credited styled run will confirm visually). Prior: [v0.26.0](LIVE_VERIFICATION_v0.26.0.md) (**image i2i by generated-image UUID proven live e2e GREEN** — the existing asset is *selected in place* by its media UUID, no duplicate upload, producing an i2i output that references it; plus generated-image `display_name` capture, credited @C1ph3r404). Prior: [v0.25.0](LIVE_VERIFICATION_v0.25.0.md) (**#237 remote-UUID i2v proven live e2e GREEN** — a real 8s 720×1280 interpolation from a generated image's UUID, after the picker-search attach was reworked to local upload; #240 home-`.env` matrix live; plus two silent-failure guards root-caused live: video-as-image download rejection and rejected-upload fail-loud). Prior: [v0.24.0](LIVE_VERIFICATION_v0.24.0.md) (--project parity across CLI + MCP — automated coverage; live gen auth-gated). Prior: [v0.23.0](LIVE_VERIFICATION_v0.23.0.md) (MCP generation wiring proven live end-to-end — tool→worker→real Flow REST→structured-error translation; final image-write blocked by an expired session, environmental. #222 macOS fix reporter-verified e2e on Apple Silicon). Prior: [v0.22.0](LIVE_VERIFICATION_v0.22.0.md) (Tools framework — automated/CI coverage complete; credit/key-gated live feature run is a pending owner gate, see the doc) · [v0.21.0](LIVE_VERIFICATION_v0.21.0.md) (MCP server over stdio + HTTP/SSE — credit-free protocol handshake; 4 tools / 2 prompts / 3 resources confirmed on stdout) · [v0.20.1](LIVE_VERIFICATION_v0.20.1.md) · [v0.20.0](LIVE_VERIFICATION_v0.20.0.md) · [v0.19.0](LIVE_VERIFICATION_v0.19.0.md) · [v0.18.0](LIVE_VERIFICATION_v0.18.0.md) · [v0.17.0](LIVE_VERIFICATION_v0.17.0.md) · [v0.16.0](LIVE_VERIFICATION_v0.16.0.md) · [v0.13.0](LIVE_VERIFICATION_v0.13.0.md) · [v0.12.0](LIVE_VERIFICATION_v0.12.0.md) · [v0.11.0](LIVE_VERIFICATION_v0.11.0.md) · [v0.10.0](LIVE_VERIFICATION_v0.10.0.md) · [v0.9.1](LIVE_VERIFICATION_v0.9.1.md) · [v0.9.0](LIVE_VERIFICATION_v0.9.0.md) · [v0.8.1](LIVE_VERIFICATION_v0.8.1.md) · [v0.7.0](LIVE_VERIFICATION_v0.7.0.md)
+**What was actually live-verified for the latest release?** → latest: [LIVE_VERIFICATION_v0.54.0](LIVE_VERIFICATION_v0.54.0.md) (**close-browser guidance #470, character-composer focus fix, overlay/jitter hardening, and SonarCloud-S2245 RNG swap live-verified 2026-08-12 on profile `ffroliva`, ZERO credits** — the new #470 login wording printed verbatim and the manual-close → `auth_flow_session_verified` round-trip worked end-to-end; a Classic-UI `t2i` produced a real 1024×1024 JPEG [`ff d8 ff e0`] exercising overlay dismissal + jitter + image-composer submit + mode controller, and a credit-free `character create` [entity `da3cd647…`, face 1376×768 JPEG] exercised the character-composer path directly [`entering_character_editor` → `prompt_submitted` → `entity_patched` → `completed`]; the first `t2i` hit the known agentic-cohort video-for-image condition [correctly refused via `WireFormatError`, tracked by #299, not a regression] and Classic UI worked around it. First mainline release after the orphaned v0.53.0/v0.53.1 tags — CHANGELOG restores honest 0.53.x sections). Prior: [LIVE_VERIFICATION_v0.52.0](LIVE_VERIFICATION_v0.52.0.md) (**batch DAG references #317, character entity provenance recording #402, and video duration selector cascade fix #451 live-verified 2026-08-05** — 3/3 live E2E tests in `tests/e2e/test_entity_provenance_e2e.py` passed clean on profile `ffroliva`, 5-layer evidence ledger verified). Prior: [v0.51.0](LIVE_VERIFICATION_v0.51.0.md) (**dependency/security/test-correctness release verified 2026-08-05, 1 Imagen credit + 1 Veo attempt** — the only change needing live evidence was the `playwright` bound raise `>=1.59.0,<1.60.0` → `>=1.61.0,<1.62.0`, and the 2026-08-03 regression does NOT reproduce on 1.61.0: a live i2v drove `image_uploaded status=200` → `frame_attached` → `generate_captured status=200` with `startImage` parsed, where **1.62.0 hung silently at exactly that upload step**; live i2i local-ref attach passed outright [96.7s]. **1.62.0 stays excluded**, never root-caused. `-m e2e_auth` 16/16, daemon e2e 1 passed [MCP over Streamable HTTP — the first live coverage that test has ever had, it was structurally incapable of passing before], MCP Tasks 5/5 layers, 2923 offline tests, `pip-audit --all-extras` clean. **No finished mp4 is claimed** — Flow's own generation returned `PUBLIC_ERROR_VIDEO_GENERATION_TIMED_OUT`, server-side. Defect found and recorded not omitted: Flow dropped the duration-tab UI so **`--duration` is currently broken for i2v**, A/B-proven pre-existing on BOTH 1.59.0 and 1.61.0 — [#451](https://github.com/ffroliva/gflow-cli/issues/451)). Prior: [v0.50.0](LIVE_VERIFICATION_v0.50.0.md) (**MCP Tasks extension SEP-2663, non-blocking task handle responses, CLI/MCP output path hardening -o/--output, and Telegram PR triage alert resilience verified 2026-08-04** — 5-layer evidence ledger note generated at `tmp/live-verify/mcp-tasks.md` confirming instant task handle return, working status queries, and cancelled status DB state transitions; full 118-test suite green). Prior: [v0.49.0](LIVE_VERIFICATION_v0.49.0.md) (**#125 `omni-flash` start-frame I2V video generation, Playwright dependency upper-bound pin `>=1.59.0,<1.60.0`, submission stage watchdog timeout, and #404 count-tab fail-closed logic verified 2026-08-03** — 0-credit wire intercept captured `batchAsyncGenerateVideoStartImage` with `startImage` bound; live costed generation on `omni-flash` I2V produced a 1.2MB valid mp4 video asset on `ffroliva`; full 2902-test release suite green). Prior: [v0.48.0](LIVE_VERIFICATION_v0.48.0.md) (**#411 predictable `-o`/`--output` paths, offline-verified 2026-08-02, ZERO credits** — exact single-target local writes + `_1`/`_2` multi-count **image** suffixes proven by the CLI suite; the pre-tag council audit [RED, then resolved] re-scoped the release to the truthful surface: a dead MCP `output` param was deleted [worker queue never read it — #414], cloud `-o` targets and video multi-count suffixes are follow-ups [#415], and two tautological video-path tests are recorded as weak evidence rather than claimed as proof; storage-layer S3 writes verified against Docker MinIO at the layer level only; 2888-test release tree green; the live-Flow run aborted pre-generation on an expired denon82 session [`AuthExpiredError` 401, correct RFC-7807 contract + remediation hint, $0] — recorded, not omitted; live re-run queued for the next authenticated session). Prior: [v0.47.0](LIVE_VERIFICATION_v0.47.0.md) (**MCP SDK 2.0.0 migration + dual-era protocol + #402 entity provenance recording live-verified 2026-08-01, ZERO credits** — stdio JSON-RPC initialization handshake verified over subprocess transport, `resolve-drift` CI job added to prevent dependency drift, and 121 unit tests in `tests/data` passed). Prior: [v0.46.1](LIVE_VERIFICATION_v0.46.1.md) (**#404 classic count-setter fix live-verified 2026-07-31 on the reporter's exact environment, ZERO credits** — credit-free DOM recon proved Flow renamed the count-tab labels to uniform `xN` [`x1` replaces `1x`, image AND video rows], that `nth(0)` of the old filtered set landed on the selected `x2` and actively moved the count AWAY from the target, and that a digit-keyed click on `x1` converges instantly and persists without a Save commit; the new classic e2e then passed live [count=1 against a fresh project displaying 2 → exactly 1 image; count=2 early-exit → exactly 2], and the #313 agentic count-enforcement e2e passed 4/4 [counts 1–4 vs deliberately mismatched sticky defaults, magic-byte-verified downloads] after its stale setup was repaired [Flow now loads the editor in classic mode — the setup needed an Agent-pill toggle before the `tune` icon; test-only fix, no product code implicated]; the video output-count cascade carries a recorded waiver [full drive costs Veo credits; its exact selector shape was exercised live in the spike's cleanup step]). Prior: [v0.46.0](LIVE_VERIFICATION_v0.46.0.md) (**#387 provider-agnostic prompt-tools transport live-verified 2026-07-28, ZERO credits** — a four-probe capability spike passed 4/4 against BOTH a non-Google gateway and Google's OpenAI-compat endpoint, including a *semantic* multimodal check [a generated solid-red JPEG round-tripped and the model answered 'Red'], which is what proved the native Gemini `inlineData` path could be deleted outright rather than kept as a fallback; the real CLI then returned `was_expanded: true` on gateway/gateway-no-model/Google-default, emitted the loud one-time notice + no-op for the removed `GFLOW_CLI_GEMINI_API_KEY`, and made NO network call when unconfigured. Live verification caught two defects a fully green 2849-test offline suite missed: Google's compat endpoint has no server-side default model [`400 "model is not specified"`, silently no-oping the DEFAULT configuration], and the first redirect guard did not actually work because `build_opener()` installs its default handlers regardless of arguments, so `Authorization` would still have been replayed cross-host. `/code-review high` is recorded as NOT run — it is user-triggered only). Prior: [v0.45.0](LIVE_VERIFICATION_v0.45.0.md) (**#393 UUID `--ref` binding + #395 character-entity binding live-verified 2026-07-27/28** — a controlled A/B on identical input flipped `--ref` from exit 9 to exit 0 via the catalog-backed upload fallback, with the anti-silence guard [an unbindable ref aborts non-zero and writes no image] pinned by a live e2e; picker DOM + HAR showed Flow indexes a short auto-caption, NOT the prompt, so a prompt-derived search hint was dropped rather than shipped on assumption; a CDP spike against Flow's own character flow then captured the `entityContext` contract that gflow's generations were missing, root-causing #395 to two gflow defects [overlay dismissal Escaping Flow's own composer, and a character-route bounce that sent the prompt to the project composer] — both fixed, both character e2e tests now pass live, including `--format-prompt` [#383], with a read-back confirming the character carries its name and `thumbnail_media_id`). Prior: [v0.44.0](LIVE_VERIFICATION_v0.44.0.md) (**#381 dual-side project management & Playwright Chromium transport live-verified 2026-07-26** — real-editor project creation + rename tRPC server/DOM sync, prompt slugging, `gflow project` subcommand family, HTTP 429 adaptive backoff, and full E2E test suite). Prior: [v0.43.0](LIVE_VERIFICATION_v0.43.0.md) (**private incident diagnostics live-verified 2026-07-23, 1 Veo credit**). Prior: [v0.42.0](LIVE_VERIFICATION_v0.42.0.md) (**#361 production-readiness live-gaps confirmed live 2026-07-21, 1 Veo credit** — a real `veo-lite` t2v through the worker path persisted a real `remote_started` checkpoint [observer→checkpoint→DB chain confirmed], crash-recovery classified the task `indeterminate` without resubmitting, and mid-launch cancellation released the `ProfileLease` credit-free; full evidence in [v0.40.0-production-readiness § "2026-07-21 live follow-up"](LIVE_VERIFICATION_v0.40.0-production-readiness.md#2026-07-21-live-follow-up-0420-cycle); #359 content-safety `400`→`ContentPolicyError` classification is unit-covered, not live-triggered [would require a deliberate policy violation]; #360/#362 are docs-only, verified by the documentation gates). Prior: [v0.41.0](LIVE_VERIFICATION_v0.41.0.md) (**#357 production-readiness hardening — stale-session fail-fast, free image gen, paid veo-lite T2V all proven live**). Prior: [v0.40.0](LIVE_VERIFICATION_v0.40.0.md) (**#344 `@`-mention asset-tagging resolution proven live e2e, ~2 Imagen credits** — `gflow character create` staged a real taggable character [bound reference images, `has_reference_images` guard], the e2e test's documented propagation wait cleared Flow's `projectInitialData` lag, and `gflow image t2i "@Zoro …"` passed with returncode 0, a real generated image, and the correctly de-tagged prompt in the catalog — no code changes needed, `resolve_and_apply` as merged in #344 already works end-to-end; pre-existing unrelated issue #174 [Flow media-library A/B flake] can still intermittently block generation regardless of this feature — recorded, not omitted). Prior: [v0.39.0](LIVE_VERIFICATION_v0.39.0.md) (**#341 failed-generation persistence proven live** — a real `image t2i` on denon82 failed at the wire [`batchGenerateImages` HTTP 400 → `WireFormatError`] and the record-on-catch funnel wrote the **first-ever `failed` row** into the production catalog [`error_type=wire-format`, redacted `error_detail`, browsable via `gflow data list errors`], **$0**; the remaining paid paths [video/chain/batch/movie/worker] are funnel-regression-tested, not force-failed live [would cost credits], and the status-aware `succeeded=false` video-poll fix is unit-locked — recorded, not omitted). Prior: [v0.38.1](LIVE_VERIFICATION_v0.38.1.md) (**#338 agentic-pin recovery proven live ON the pinned account** — run 1 caught the composer-render race the unit suite could not [`ensure_media_incomplete` +115 ms, toggle never reached]; the readiness-wait fix landed and run 2 recovered classic on denon82 after ~2h of pin [`ui_driver.bound mode=classic`, real 768×1376 JPEG, exact prompt adherence]; reload branch unit-locked, live trigger not observed this cycle — recorded, not omitted). Prior: [v0.38.0](LIVE_VERIFICATION_v0.38.0.md) (**#332 mode control + #314 i2i ref dedup proven live e2e** — `image t2i --ui-mode classic` recovered agentic→classic in a real run [`ui_mode.attempt_exit_agent` → full classic chain, 1376×768 JPEG]; two identical `image i2i` runs proved the dedup contract [run 1: picker miss → `image_uploaded`; run 2: `reference_deduped_by_filename`, ZERO uploads]; `video t2v` veo-fast [`MEDIA_GENERATION_STATUS_SUCCESSFUL`, 15.2 MB `ftyp` mp4]; autopilot #333 recorded deferred-with-reason [hermes-ops deployment staged separately]). Prior: [v0.37.0](LIVE_VERIFICATION_v0.37.0.md) (**viewport 1920×1080 + agentic count enforcement + FIPS-safe SAPISIDHASH proven live e2e** — `image t2i` [classic cohort, real 1376×768 JPEG] and `video t2v` veo-fast [`MEDIA_GENERATION_STATUS_SUCCESSFUL`, 11.5 MB `ftyp` mp4] both succeeded end-to-end at the new 1920×1080 viewport with every selector resolving [no drift]; SAPISIDHASH digest verified byte-identical [FIPS-safe `usedforsecurity=False`], auth unchanged). Prior: [v0.36.0](LIVE_VERIFICATION_v0.36.0.md) (**#316 diagnostic tooling live-verified against real Flow traffic** — `GFLOW_CLI_HAR_PATH` captured a real 54-entry HAR [real Flow domains, 34× 200s, and a genuine 401 the tool exists to diagnose]; `GFLOW_CLI_DEBUG_TRACEBACK`'s typed-error scoping confirmed live [console output identical with/without the flag on a `GFlowError`]; unhandled-path behavior covered by 6 new tests (30 total across both files) + SDD task review; #312 reference-entity-smuggling fix covered by its own dedicated e2e test). Prior: [v0.35.0](LIVE_VERIFICATION_v0.35.0.md) (**#305-follow-up multimodal reverse-engineering & storyboard tool proven live e2e** — `gflow tools run storyboard` successfully deconstructed a cyberpunk scene into a cohesive 4-panel sequence; `reverse-engineer` tool extracts frames from video streams/URLs using `claude-video`'s `watch.py` and produces structured prompt components using Gemini multimodal models; dynamic token budgeting and strict Pyright types verified via test suite). Prior: [v0.34.0](LIVE_VERIFICATION_v0.34.0.md) (**#299 `--ui-mode` gate proven live, ZERO credits** — drove `get_ui_driver` directly over 4 real editor loads, never generated: bidirectional switch works [classic→agentic via `_force_agent_mode`, agentic→classic via `_exit_agent_mode`, each DOM-reverified], `agentic`-when-already-agentic is idempotent, and a timed-out toggle → **exit-28 fail-fast before submission**; deprecation behavior-change [`prefer_classic`→exit 28] confirmed). Prior: [v0.33.0](LIVE_VERIFICATION_v0.33.0.md) (**#287-follow-up picker fix proven live on the original repro** — the v0.32.1-failing `i2v <uuid> --project <crowded>` command now exits 0 with a real 3.0 MB `ftyp` mp4, `frame_ref_attached` + `video_saved`, zero uploads; **#241 jitter live-verified on both paths** — flag `--jitter 4-6` → `batch_jitter_sleep seconds=4.98`, and the new 0.5–1.5 s default → `seconds=1.47`, each firing exactly once between two t2i submissions; environment note: the denon82 cohort is now natively agentic and round-2 generations hit the pre-existing #281 `MediaAttributionError`, not a regression). Prior: [v0.32.1](LIVE_VERIFICATION_v0.32.1.md) (**#293 teardown fix live-verified credit-free** — error-path aborts leave ZERO chrome.exe holding the profile [3× leak reproduced pre-fix same day]; force-close fallback + ProfileLockedError translation unit-pinned; #283 wave logic fixes mock-verified to exact call counts). Prior: [v0.32.0](LIVE_VERIFICATION_v0.32.0.md) (**#287 UUID i2v frames proven live e2e** — an in-project asset selected by media UUID in the frame-slot dialog [frame_ref_attached, startImage bound on the wire, ZERO upload events] produced a real 720×1280 interpolation, resolving the #237 negative prior; foreign UUID → exit 9 pre-generation with screenshot; **#288 duration fail-fast proven live** — exit 23 pre-submit saved a 10-credit generation, and the debug screenshot confirmed the duration control is ABSENT from the cohort's settings popover [locale hypothesis refuted]; upload-400 exit-27 repro deferred-with-reason — the original rejected JPEG no longer exists). Prior: [v0.31.0](LIVE_VERIFICATION_v0.31.0.md) (**#281 wrong-media defenses live-verified in BOTH directions, credit-free** — correct attribution downloads with real wire metadata [seed, 768×1376]; the incident-class lazy-render ambiguity reproduced live and hard-stopped with `MediaAttributionError` exit 26 naming every candidate UUID, no file written; **#282 two-UUID-ref i2i proven e2e** — both freshly generated portrait UUIDs picker-selected into one scene with both canonical identities visually transferred). Prior: [v0.30.0](LIVE_VERIFICATION_v0.30.0.md) (**agentic-cohort image path + both fixes live-verified credit-free** — native 768×1376 stills, i2i-chain + canonical-ref identity hold; the character-create `DataIntegrityError` and the `prefer_classic` WARNING were observed live then fixed [recorder idempotent + INFO log]; MCP video-param feature deferred-with-reason on Veo credits, per the stills-only goal). Prior: [v0.29.0](LIVE_VERIFICATION_v0.29.0.md) (**persistent `gflow instructions` CRUD proven live e2e GREEN, credit-free** — 11-command sequence on a fresh project: add incl. `--ref <image-UUID>` attach, list/enable/disable, master toggle both ways, TOML `apply` destructive full-sync with `enabled=false` honored, `rm` with server card-id preserved across mutations; movie brief-sync code-path-verified via the same primitives, deferred-with-reason on credits; MCP `gflow_instructions_*` tools are thin adapters over the live-verified primitives). Prior: [v0.28.0](LIVE_VERIFICATION_v0.28.0.md) (**agent instructions `-i` now steer agentic generation — crayon e2e GREEN**: a style-neutral prompt + an enabled "crayon" card produced an unmistakable crayon drawing through the real transport; root causes were conversational phrasing + the `project_brief.enabled` master switch). Prior: [v0.27.1](LIVE_VERIFICATION_v0.27.1.md) (**v0.27.0 release follow-up fixes and documentation sync**). Prior: [v0.27.0](LIVE_VERIFICATION_v0.27.0.md) (**`[style]` block with named variants + prompt-aware resume — credit-free CLI verification**: real `movie run --dry-run` 3-way ledger proving resolved-style display, `skip (done)` on matching `style_hash`, and `re-run (style changed)` after a style edit; composition is local, wire path untouched, so no Veo credits spent — first credited styled run will confirm visually). Prior: [v0.26.0](LIVE_VERIFICATION_v0.26.0.md) (**image i2i by generated-image UUID proven live e2e GREEN** — the existing asset is *selected in place* by its media UUID, no duplicate upload, producing an i2i output that references it; plus generated-image `display_name` capture, credited @C1ph3r404). Prior: [v0.25.0](LIVE_VERIFICATION_v0.25.0.md) (**#237 remote-UUID i2v proven live e2e GREEN** — a real 8s 720×1280 interpolation from a generated image's UUID, after the picker-search attach was reworked to local upload; #240 home-`.env` matrix live; plus two silent-failure guards root-caused live: video-as-image download rejection and rejected-upload fail-loud). Prior: [v0.24.0](LIVE_VERIFICATION_v0.24.0.md) (--project parity across CLI + MCP — automated coverage; live gen auth-gated). Prior: [v0.23.0](LIVE_VERIFICATION_v0.23.0.md) (MCP generation wiring proven live end-to-end — tool→worker→real Flow REST→structured-error translation; final image-write blocked by an expired session, environmental. #222 macOS fix reporter-verified e2e on Apple Silicon). Prior: [v0.22.0](LIVE_VERIFICATION_v0.22.0.md) (Tools framework — automated/CI coverage complete; credit/key-gated live feature run is a pending owner gate, see the doc) · [v0.21.0](LIVE_VERIFICATION_v0.21.0.md) (MCP server over stdio + HTTP/SSE — credit-free protocol handshake; 4 tools / 2 prompts / 3 resources confirmed on stdout) · [v0.20.1](LIVE_VERIFICATION_v0.20.1.md) · [v0.20.0](LIVE_VERIFICATION_v0.20.0.md) · [v0.19.0](LIVE_VERIFICATION_v0.19.0.md) · [v0.18.0](LIVE_VERIFICATION_v0.18.0.md) · [v0.17.0](LIVE_VERIFICATION_v0.17.0.md) · [v0.16.0](LIVE_VERIFICATION_v0.16.0.md) · [v0.13.0](LIVE_VERIFICATION_v0.13.0.md) · [v0.12.0](LIVE_VERIFICATION_v0.12.0.md) · [v0.11.0](LIVE_VERIFICATION_v0.11.0.md) · [v0.10.0](LIVE_VERIFICATION_v0.10.0.md) · [v0.9.1](LIVE_VERIFICATION_v0.9.1.md) · [v0.9.0](LIVE_VERIFICATION_v0.9.0.md) · [v0.8.1](LIVE_VERIFICATION_v0.8.1.md) · [v0.7.0](LIVE_VERIFICATION_v0.7.0.md)
 **"Was the playwright upper bound raised, and on what evidence?"** → [LIVE_VERIFICATION_playwright_1.61](LIVE_VERIFICATION_playwright_1.61.md) (**`>=1.59.0,<1.60.0` → `>=1.61.0,<1.62.0` verified 2026-08-05, 1 Imagen + 1 Veo credit** — the 2026-08-03 regression that motivated the bound does NOT reproduce on 1.61.0: a live `i2v` drove the full chain with `image_uploaded status=200` → `frame_attached` → `generate_captured status=200` with `startImage` parsed, where **1.62.0 hung silently at exactly that upload step**; live `i2i` local-ref attach passed outright [96.7s]. **1.62.0 stays excluded** — never root-caused. Also found, A/B-proven pre-existing on BOTH 1.59.0 and 1.61.0: Flow dropped the duration-tab UI, so **`--duration` is currently broken for i2v** — `UiSelectorDriftError` fires correctly [refusing rather than silently accepting Flow's default, #288] but the selector needs re-deriving; that is why the i2v proof was driven via the CLI without `--duration`, and Flow's own server-side `PUBLIC_ERROR_VIDEO_GENERATION_TIMED_OUT` means no finished mp4 is claimed).
 **"Where is the reverse-engineered wire protocol for a feature?"** → the `*_RECON.md` design docs: [CHARACTER_RECON](CHARACTER_RECON.md) (Flow character entity protocol), [IMAGE_UPSCALE_RECON](IMAGE_UPSCALE_RECON.md) (`/v1/flow/upsampleImage` wire). Naming convention: one `<FEATURE>_RECON.md` per reverse-engineered surface, kept as the durable spec after the feature ships. Pre-capture research recons: [ASSET_TAGGING_RECON](ASSET_TAGGING_RECON.md) (`@`-mention asset tagging — shipped in v0.40.0, see [LIVE_VERIFICATION_v0.40.0](LIVE_VERIFICATION_v0.40.0.md)).
 **"What was live-verified for the data layer (PR #58)?"** → [LIVE_VERIFICATION_data_layer](LIVE_VERIFICATION_data_layer.md) — 1 Imagen + 1 Veo credit on denon82, 6-layer ledger (file + magic + Pillow + DB rows + CLI round-trip + structlog)

gflow_cli-0.54.0/docs/LIVE_VERIFICATION_v0.54.0.md

@@ -0,0 +1,67 @@
+# Live Verification — v0.54.0 (Close-Browser Guidance, Composer/Overlay Hardening, Jitter RNG)
+
+**Date:** 2026-08-12
+**Profile:** `ffroliva` (`ffroliva@gmail.com`) — real-Chrome auth strategy, running the **0.54.0** build from `chore/release-v0.54.0`.
+**Credits spent:** 0 — `image t2i` and `character create` (face) are credit-free; no Veo generation was driven.
+
+## 1. What This Document Verifies
+
+v0.54.0 ships **no generation feature**; its user-facing surface is a login-message reword (#470), a character-composer focus fix, overlay/watermark selector hardening (#403 refinement), a cryptographic RNG swap for interaction jitter (SonarCloud S2245), and dependabot guards for the playwright/patchright drivers (#465). This document records **live exercise of every changed code path** against real Flow on the 0.54.0 build. Because the changes live in the shared UI-automation submit/overlay/jitter machinery, two credit-free generations (a Classic-UI `t2i` and a `character create`) exercise all of them end-to-end.
+
+## 2. 5-Layer Verification Ledger
+
+| Layer / Change | Recorded Live Evidence | Result |
+|---|---|---|
+| **#470 — close-browser guidance** | The reworded step 4 printed **verbatim** during the live `auth login` on the 0.54.0 build; the manual close then drove `Browser closed. Verifying Flow session…` → `auth_flow_session_verified` (`ffroliva@gmail.com`) → `[OK] Flow session verified`. End-to-end on the real-Chrome path. | 🟢 PASS |
+| **1. File count (t2i)** | `1/1 succeeded` — Classic-UI `image t2i` | 🟢 PASS |
+| **2. Magic bytes (t2i)** | `ff d8 ff e0` (JPEG/JFIF) | 🟢 PASS |
+| **3. Dimensions (t2i)** | Pillow: `JPEG (1024, 1024) RGB`, 776 KB — [`0754aa44…_1.jpg`](file:///C:/Users/ffrol/Downloads/gflow-cli/images/2026-08-12/0754aa44-b5fd-42b8-85ee-e58e0fb25175_1.jpg) | 🟢 PASS |
+| **4. Structlog invariants (t2i)** | `ui_driver.ui_mode.attempt_exit_agent` → `ui_automation.aspect_ratio_set` → `ui_automation.prompt_submitted` → `ui_automation.batch_response_captured` (all `cli_version 0.54.0`) — exercises overlay dismissal, interaction jitter, image-composer submit, and the mode controller | 🟢 PASS |
+| **Character-composer focus fix** | `character create` → entity `da3cd647-6ea2-4075-af8b-9404c9897c37` (`LiveVerify054`); composer path `ui_automation.entering_character_editor` → `character_editor_ready` → `character_model_selected` → `prompt_input_found` → `prompt_submitted` → `character_create.face_done` → `entity_patched` → `completed`, no errors | 🟢 PASS |
+| **5. Character face asset** | `ff d8 ff e0` — Pillow `JPEG (1376, 768) RGB`, 556 KB — [`character_da3cd647…_slot0.jpg`](file:///C:/Users/ffrol/Downloads/gflow-cli/characters/2026-08-12/character_da3cd647-6ea2-4075-af8b-9404c9897c37_slot0.jpg) | 🟢 PASS |
+| **Jitter RNG (SonarCloud S2245)** | `_jitter_ms` interaction jitter ran during **both** live generations with no error (internal swap to `secrets.SystemRandom`); primary coverage via the SonarCloud gate + offline suite | 🟢 PASS (path exercised; Sonar/unit-covered) |
+| **Overlay/watermark hardening (#403 refinement)** | No blocking release overlay appeared on the fresh session, so the pure-structural selectors ran without a live modal to dismiss this cycle; behavior is unit-covered (`tests/api/transports/test_ui_automation.py`) and the underlying #403 dismissal was live-verified in v0.53.0 | 🟢 PASS (unit-covered; no live modal this cycle) |
+| **#465 / patchright dependabot guards** | Not a live-Flow feature — enforced by `tests/test_playwright_pin.py` (`test_dependabot_ignores_playwright_minor_bumps`, `test_dependabot_ignores_driver_engine_bumps`) | 🟢 PASS (test-covered) |
+
+## 3. Execution Log Trace (Verbatim Excerpt)
+
+Login (#470 message → session verified):
+
+```
+4. When you're finished, simply close the Chrome window — that's how you let gflow know you're done.
+   gflow then verifies your Flow session automatically; there's nothing else to do.
+{"strategy": "chrome", "source": "chrome", "user_email": "ffroliva@gmail.com", "event": "auth_flow_session_verified", "cli_version": "0.54.0", ...}
+[OK] Flow session verified (ffroliva@gmail.com).
+```
+
+Classic-UI `image t2i` (overlay / jitter / composer / mode controller):
+
+```
+{"event": "ui_driver.ui_mode.attempt_exit_agent", "cli_command": "image t2i", "cli_version": "0.54.0", ...}
+{"event": "ui_automation.aspect_ratio_set",       "cli_command": "image t2i", "cli_version": "0.54.0", ...}
+{"event": "ui_automation.prompt_submitted",        "cli_command": "image t2i", "cli_version": "0.54.0", ...}
+{"event": "ui_automation.batch_response_captured", "cli_command": "image t2i", "cli_version": "0.54.0", ...}
+```
+
+`character create` (character-composer path):
+
+```
+{"event": "ui_automation.prompt_submitted",        "cli_command": "character create", "cli_version": "0.54.0", ...}
+{"entity_id": "da3cd647-...", "media_id": "bc36a2ea-...", "event": "character_create.face_done", ...}
+{"entity_id": "da3cd647-...", "event": "character_create.entity_patched", ...}
+{"entity_id": "da3cd647-...", "name": "LiveVerify054", "event": "character_create.completed", ...}
+```
+
+## 4. User-Confirmable Outputs
+
+- **Image:** `~/Downloads/gflow-cli/images/2026-08-12/0754aa44-b5fd-42b8-85ee-e58e0fb25175_1.jpg` — a serene mountain lake at dawn, 1024×1024.
+- **Character face:** `~/Downloads/gflow-cli/characters/2026-08-12/character_da3cd647-6ea2-4075-af8b-9404c9897c37_slot0.jpg` — a red-fox portrait, 1376×768.
+- **Character entity:** `da3cd647-6ea2-4075-af8b-9404c9897c37` (`LiveVerify054`) in project `e00291af-a329-42de-ab2a-1dbcc71137fc` (browsable via `gflow character show --id da3cd647-…`).
+
+## 5. Known Condition (documented, not a blocker)
+
+The **first** `t2i` attempt hit the known Flow **agentic-UI cohort** behavior — the conversational agent returned *video* content for an image request. gflow correctly **detected and refused it** (`WireFormatError`, first-class guard at `api/client.py:1572`, remediation: `GFLOW_CLI_PREFER_CLASSIC=1`) rather than saving a wrong file. This is the pre-existing behavior tracked by open issue **#299** (UI-mode reliability / agentic driver hardening) — **not a 0.54.0 regression**. Forcing Classic UI worked around it, and the successful runs above followed. Recorded, not omitted.
+
+## 6. Release Reconciliation Note
+
+v0.54.0 is the first mainline release after **v0.53.0 / v0.53.1**, which were published as GitHub releases (tags) but whose commits never reached `main`/`develop`. Their content (#315 driver-delay humanization, #403 overlay detection) was already present in `develop` as patch-equivalents; this release's CHANGELOG restores honest `[0.53.0]` / `[0.53.1]` sections and supersedes the orphaned tags. See `CHANGELOG.md`.

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/docs/SECURITY.md

@@ -259,10 +259,15 @@ Controls:
 
 - **Optional-only** — never in the base `dependencies`; the default install pulls
   neither the package nor its driver.
-- **Exact-pinned** (`patchright==1.60.1`) so a release cannot silently move the
+- **Exact-pinned** (`patchright==1.61.2`) so a release cannot silently move the
   browser binary underneath you.
 - **Security-review-required on bump** — a Patchright version change is treated
-  like a browser-binary change, NOT a routine dependabot auto-merge.
+  like a browser-binary change, NOT a routine dependabot auto-merge. The exact
+  pin alone does **not** enforce this: Dependabot's `uv` ecosystem rewrites a
+  constraint standing in an update's way rather than respecting it (PR #465 did
+  exactly that to playwright's upper bound). The enforcement is the explicit
+  `ignore` entry in `.github/dependabot.yml`, guarded by
+  `tests/test_playwright_pin.py::test_dependabot_ignores_driver_engine_bumps`.
 - Default engine is `playwright` (Microsoft, security-reviewed); patchright is an
   experiment you opt into per the trade-off above.
 

gflow_cli-0.54.0/docs/superpowers/plans/2026-08-06-ui-mode-policy/PLAN.md

@@ -0,0 +1,98 @@
+# Per-Command UI Mode Policy Implementation Plan
+
+> **For agentic workers:** Run `/gflow:status --feature ui-mode-policy` to find the next unchecked task. Implement one task at a time. Run `/gflow:check` before every commit.
+
+**Goal:** Add `--ui-mode auto|classic|agentic` CLI option, `GFLOW_CLI_UI_MODE` setting, and strict policy validation before generation submission to fail fast with `UiModeUnavailableError` (exit 28) when requested UI mode doesn't match the mounted cohort.
+
+**Predict verdict:** GO — confidence 10/10
+
+---
+
+## File structure
+
+### Modified files
+```
+src/gflow_cli/config.py
+  Add ui_mode setting ("auto" | "classic" | "agentic").
+src/gflow_cli/api/transports/drivers/factory.py
+  Enforce strict mode policy check in bind_ui_driver, raising UiModeUnavailableError.
+src/gflow_cli/cli_image.py, cli_video.py, cli_run.py
+  Add --ui-mode Click option.
+src/gflow_cli/mcp/tools.py (and server definitions)
+  Mirror ui_mode tool argument for CLI-MCP parity.
+tests/api/transports/test_driver_factory.py
+  Unit tests for strict UI mode policy validation.
+tests/mcp/test_cli_parity.py
+  CLI-MCP parity test assertions.
+CHANGELOG.md
+  Unreleased section update.
+```
+
+---
+
+## Task 1 — Config & Settings Extension
+
+**What:** Add `ui_mode: Literal["auto", "classic", "agentic"] = "auto"` setting to `Settings` in `config.py`.
+
+**Files:**
+- `src/gflow_cli/config.py`
+- `tests/test_config.py`
+
+**Steps:**
+- [ ] Add `ui_mode` setting in `config.py` with validation for `auto`, `classic`, `agentic`.
+- [ ] Add unit test in `tests/test_config.py`.
+
+---
+
+## Task 2 — Driver Factory Policy Enforcement
+
+**What:** Update `bind_ui_driver` in `src/gflow_cli/api/transports/drivers/factory.py` to check the configured/requested `ui_mode`. If `ui_mode != "auto"` and `detected_mode != ui_mode`, raise `UiModeUnavailableError`.
+
+**Files:**
+- `src/gflow_cli/api/transports/drivers/factory.py`
+- `tests/api/transports/test_driver_factory.py`
+
+**Steps:**
+- [ ] Implement `ui_mode` parameter in `bind_ui_driver`.
+- [ ] Raise `UiModeUnavailableError` on mismatch.
+- [ ] Add unit tests in `tests/api/transports/test_driver_factory.py`.
+
+---
+
+## Task 3 — CLI & MCP Option Symmetry
+
+**What:** Add `--ui-mode` Click option to generation commands and update MCP tool definitions.
+
+**Files:**
+- `src/gflow_cli/cli_image.py`
+- `src/gflow_cli/cli_video.py`
+- `src/gflow_cli/cli_run.py`
+- `src/gflow_cli/mcp/tools/`
+- `tests/mcp/test_cli_parity.py`
+
+**Steps:**
+- [ ] Add `--ui-mode` Click option to `gflow image t2i`, `gflow image i2i`, `gflow video t2v`, `gflow video i2v`, `gflow run`.
+- [ ] Update MCP tool definitions.
+- [ ] Verify `tests/mcp/test_cli_parity.py` passes.
+
+---
+
+## Task 4 — Quality Gates & Pre-Commit Validation
+
+**What:** Run the Impeccable Routine (`/gflow:check`).
+
+**Steps:**
+- [ ] `uv run python scripts/ci/check_repo_hygiene.py`
+- [ ] `uv run python scripts/ci/check_doc_links.py`
+- [ ] `uv run ruff check src tests`
+- [ ] `uv run ruff format --check src tests`
+- [ ] `uv run pyright src`
+- [ ] `uv run python -m pytest -q`
+
+---
+
+## Definition of Done
+
+- [ ] All task steps checked off
+- [ ] `/gflow:check` green
+- [ ] `CHANGELOG.md` `[Unreleased]` section updated

gflow_cli-0.54.0/docs/superpowers/plans/2026-08-06-ui-mode-policy/SCENARIO.md

@@ -0,0 +1,27 @@
+# Scenario: Issue #299 — Per-Command UI Mode Policy
+
+## Coverage Map
+- **Active Dimensions:**
+  - **D11 (Input validation & boundaries):** Valid `--ui-mode` values (`auto`, `classic`, `agentic`). Invalid values raise `ConfigurationError`.
+  - **D6 (CLI & MCP Parity):** Every CLI generation command with `--ui-mode` must have a corresponding MCP tool parameter.
+  - **D8 (Typed Error Exit Codes):** Mismatch in strict mode (`--ui-mode classic` on agentic DOM, or `--ui-mode agentic` on classic DOM) raises `UiModeUnavailableError` with exit code 28.
+- **Skipped Dimensions:** D1, D2, D4, D5, D7, D9, D10, D12.
+
+---
+
+## Scenario Table
+
+| # | Dimension | Scenario | Severity | Expected behaviour | Test category |
+|---|---|---|---|---|---|
+| 1 | D11 Input validation | `--ui-mode invalid` passed to CLI | High | Raises `ConfigurationError` | Unit |
+| 2 | D8 Error exit code | `--ui-mode classic` when DOM is agentic | High | Raises `UiModeUnavailableError` (exit code 28) | Unit |
+| 3 | D8 Error exit code | `--ui-mode agentic` when DOM is classic | High | Raises `UiModeUnavailableError` (exit code 28) | Unit |
+| 4 | D6 Schema symmetry | `gflow image t2i --ui-mode` options | High | Mirrored in MCP `generate_image` tool schema | Parity Unit |
+
+---
+
+## Must-Cover Before Merge
+1. Add `ui_mode` setting in `config.py` with `auto`, `classic`, `agentic` literals.
+2. Implement strict UI mode enforcement in `factory.py::bind_ui_driver` raising `UiModeUnavailableError`.
+3. Add `--ui-mode` Click option across generation commands (`cli_image.py`, `cli_video.py`, `cli_run.py`) and update MCP tool definitions.
+4. Unit tests in `tests/api/transports/test_driver_factory.py` and `tests/mcp/test_cli_parity.py`.

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "gflow-cli"
-version = "0.53.0"
+version = "0.54.0"
 description = "Unofficial CLI for Google Flow — drive Veo image-to-video generations from the terminal."
 readme = "README.md"
 license = { file = "LICENSE" }
@@ -62,7 +62,7 @@ dependencies = [
 dev = [
     "pytest>=8.0.0",
     "pytest-asyncio>=0.23.0",
-    "ruff==0.16.1",
+    "ruff==0.16.2",
     "pyright>=1.1.0",
     # video-chain: the last-frame extractor (media.py) decodes with PyAV, and its
     # tests build synthetic mp4 fixtures with numpy. Declared here so the dev/CI
@@ -173,7 +173,7 @@ dev = [
     "pytest-asyncio>=1.3.0",
     "pytest-bdd>=8.1.0",
     "pytest-cov>=7.1.0",
-    "ruff==0.16.1",
+    "ruff==0.16.2",
 ]
 # Container-based integration tests — requires Docker daemon.
 # Install: uv sync --group containers

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/src/gflow_cli/__init__.py

@@ -1,3 +1,3 @@
 """gflow-cli — unofficial CLI for Google Flow."""
 
-__version__ = "0.53.0"
+__version__ = "0.54.0"

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/src/gflow_cli/api/transports/ui_automation.py

@@ -18,6 +18,7 @@ import hashlib
 import json
 import random
 import re
+import secrets
 import time
 from typing import TYPE_CHECKING, Any, cast
 from urllib.parse import urlparse
@@ -450,29 +451,20 @@ TOP_BANNER_SELECTORS: tuple[str, ...] = (
 # Close-button selectors tried after a changelog iframe or banner overlay is detected.
 # Ordered from most-specific to most-generic so a precise match wins first.
 # All are tried before the Escape fallback.
-OVERLAY_CLOSE_BUTTON_SELECTORS = (
-    # Structural & ARIA anchors (100% language-agnostic)
+OVERLAY_CLOSE_BUTTON_SELECTORS: tuple[str, ...] = (
     "[role='dialog']:has(a[href*='changelog']) button",
-    "[aria-label='Close']",
-    "[aria-label='close']",
-    "[aria-label='Dismiss']",
-    "[aria-label='dismiss']",
-    "[aria-label*='Dismiss' i]",
-    "[aria-label='Cancel']",
     "button:has(i.google-symbols:text('clear'))",
     "button:has(i:text('clear'))",
     "button:has(i.google-symbols:text('close'))",
     "button:has(i:text('close'))",
-    "[aria-label*='Got it' i]",
     "[role='dialog'] button:has(i:text('close'))",
-    "[role='dialog'] button[aria-label*='close' i]",
     "button[data-dismiss]",
 )
 
-# Detectors for the "Welcome to Google Flow" splash screen.
-WELCOME_SCREEN_SELECTORS = (
-    "div:has-text('Welcome to Google Flow')",
-    "button:has-text('See what's new')",
+# Detectors for the splash screen or welcome overlay (pure structural anchors).
+WELCOME_SCREEN_SELECTORS: tuple[str, ...] = (
+    "[role='dialog']:has(a[href*='flow'])",
+    "[role='dialog']:has(a[href*='changelog'])",
 )
 
 
@@ -573,9 +565,6 @@ def aspect_cli_from_enum(aspect: Aspect) -> str | None:
 
 # Back-compat alias — kept so any remaining internal callers and the existing
 # test imports work without a rename sweep.
-_aspect_cli_from_enum = aspect_cli_from_enum
-
-
 def _prompt_hash_stable(text: str) -> str:
     """Truncated sha256 matching image_batch._prompt_hash prefix length.
 
@@ -782,7 +771,7 @@ def _entity_ids_from_request_body(post_data: str | None) -> set[str]:
 
 
 def _jitter_ms(base_ms: int, variance: float = 0.25) -> int:
-    """Calculate a randomized delay duration bounded around base_ms.
+    """Calculate a randomized delay around `base_ms` with `±variance` spread.
 
     Adds timing entropy to browser interaction delays to break deterministic Playwright
     automation fingerprints. Returns 0 if base_ms <= 0.
@@ -790,7 +779,7 @@ def _jitter_ms(base_ms: int, variance: float = 0.25) -> int:
     if base_ms <= 0:
         return 0
     delta = int(round(base_ms * max(0.0, min(1.0, variance))))
-    return max(1, random.randint(base_ms - delta, base_ms + delta))
+    return max(1, secrets.SystemRandom().randint(base_ms - delta, base_ms + delta))  # NOSONAR
 
 
 class UiAutomationTransport(VideoGenerationMixin):
@@ -1068,7 +1057,7 @@ class UiAutomationTransport(VideoGenerationMixin):
                 loc = page.locator(close_sel).first
                 if await loc.is_visible(timeout=500):
                     await loc.click(force=True)
-                    await page.wait_for_timeout(1000)
+                    await page.wait_for_timeout(_jitter_ms(1000))
                     log.info(
                         _EVT_OVERLAY_DISMISSED,
                         selector=close_sel,
@@ -1089,7 +1078,7 @@ class UiAutomationTransport(VideoGenerationMixin):
                     loc = frame.locator(close_sel).first
                     if await loc.is_visible(timeout=500):
                         await loc.click(force=True)
-                        await page.wait_for_timeout(1000)
+                        await page.wait_for_timeout(_jitter_ms(1000))
                         log.info(
                             _EVT_OVERLAY_DISMISSED,
                             selector=close_sel,
@@ -1136,7 +1125,7 @@ class UiAutomationTransport(VideoGenerationMixin):
         # Escape fallback (regression test case: iframe present, no close button).
         try:
             await page.keyboard.press("Escape")
-            await page.wait_for_timeout(1000)
+            await page.wait_for_timeout(_jitter_ms(1000))
             log.info(
                 _EVT_OVERLAY_DISMISSED,
                 selector="<none>",
@@ -1516,7 +1505,7 @@ class UiAutomationTransport(VideoGenerationMixin):
         await input_box.press("Control+A")
         await input_box.press("Delete")
         await input_box.press_sequentially(full_prompt)
-        await page.wait_for_timeout(500)
+        await page.wait_for_timeout(_jitter_ms(500))
 
         # Readback guard — abort BEFORE submit if the typing landed in the
         # portrait box (mode switch not settled → the 2026-07-25 corruption).

{gflow_cli-0.53.0 → gflow_cli-0.54.0}/src/gflow_cli/auth/real_chrome.py

@@ -95,8 +95,11 @@ def _print_login_instructions() -> None:
         "   Signing in to Google is NOT enough; gflow needs a completed Flow app sign-in.",
     )
     _console.print(
-        "4. Then [bold yellow]CLOSE THE BROWSER[/bold yellow] — gflow verifies "
-        "the Flow session automatically.",
+        "4. When you're finished, simply [bold]close the Chrome window[/bold] — "
+        "that's how you let gflow know you're done.",
+    )
+    _console.print(
+        "   gflow then verifies your Flow session automatically; there's nothing else to do.",
     )
     _console.print("-" * 60)
     _console.print("Launching Chrome...")