@ycheung5/pi-computer-use 0.1.0 → 0.1.1

package/docs/validation.md

@@ -1,6 +1,54 @@
 # Release validation
 
-Release status: **publication pending**. Installation acceptance #2 is complete for the user-approved first-release scope; final aggregate release approval/publication remains #21. Automated and passive fixture results are evidence for their stated paths, not proof that all applications, physical input, displays or user acceptance work. The [spec](spec.md) defines requirements; this page maps them to runnable checks and pending evidence.
+Release status: **0.1.0 published on npm, 2026-11-09; 0.1.1 preparation below**. The user published the prepared package after accepting its first-release scope. Public version/latest, maintainer identity and the downloaded registry archive checksum are independently verified. #21 and product parent #1 are complete; accepted test limitations remain in force. Automated and passive fixture results are evidence for their stated paths, not proof that all applications, physical input, displays or user acceptance work. The [spec](spec.md) defines requirements; this page maps them to runnable checks and pending evidence.
+
+## 0.1.1 release preparation
+
+Prepared version: **`@ycheung5/pi-computer-use@0.1.1`**, 2026-11-10. [PR #44](https://github.com/ycheung5/pi-computer-use/pull/44) is merged as `f9a64b4`; the merged integration branch is removed locally and remotely. Package and lockfile versions are updated together. npm 0.1.0 remains the published baseline until a separate successful publication; no release tag has been created by this preparation.
+
+Release notes:
+
+- Tap standalone Ctrl/Control, Alt/Option and Shift.
+- Set optional `holdMs` for a keypress/chord or single mouse click, with release before the call settles.
+- Set optional `inputDelayMs` for explicitly selected native SendInput text; omission keeps the 50 ms default and zero removes intentional inter-unit delay.
+- Set optional `timeoutMs` on these input calls: positive overrides the execution deadline, zero disables it. When omitted, explicit hold/pacing gets a duration-aware execution budget across both request layers.
+- Retain cancellation, emergency stop, target guards, cleanup acknowledgement and non-replay. Unsupported routes and unrepresentable durations refuse before input.
+
+The [input acceptance receipt](#input-timing-and-modifier-tap-acceptance) owns Windows behavior evidence and its fixture limitations; it is reused without repeating physical tests. This release does not add broader keyboard/device coverage or promise that lower text delays work in every application. Existing first-release platform and Windows protection limits remain in force.
+
+Preparation checks passed: `npm run check` reported **214 passed, one historical-reference skip** (`/tmp/pi-cu-011-check.log`); `npm run test:package` verified the **95-entry 0.1.1 archive**, package/lock/packed-version coherence, license notices, development-file exclusion and local documentation links (`/tmp/pi-cu-011-package.log`). No runtime behavior was changed during release preparation, and no desktop or model tests were repeated.
+
+## Input timing and modifier-tap acceptance
+
+Accepted implementation scope, 2026-11-10: specification #40 and children #41/#42/#43 on `integration/input-timing-40`. Baseline `201e845`; production corrections are in `87e1f1b`, final owned-fixture corrections in `4b63389`. This is post-release source work: npm **0.1.0 remains unchanged**, with no version bump, main-branch merge or registry publication in this run. The [spec](spec.md#input-timing-and-modifier-taps) owns the behavior; [delivery](implementation-plan.md#input-timing-delivery) and [queue](tickets.md#execution-queue) own execution/status.
+
+Implemented: standalone generic Ctrl/Control, Alt/Option and Shift; per-call keypress/click holdMs; explicit native SendInput inputDelayMs with the 50 ms default retained; timeoutMs positive override/zero-disabled execution deadline and duration-aware omitted defaults across the input request layers. Safe whole milliseconds have no arbitrary product ceiling; invalid/derived-overflow values refuse before input. Long waits use representable cancellable intervals rather than overflowing platform timers. Unsupported methods/routes refuse before input. Per-call release, target/protection checks, independent stopping, cleanup uncertainty and non-replay remain binding. Broader input coverage and whole-task performance diagnosis remain outside this slice.
+
+| Evidence | Observed result and scope |
+| --- | --- |
+| TDD and portable requests | Implementers retained red/green logs under `/tmp/keypress41-*`, `/tmp/pi-cu-text42-*` and `/tmp/mouse43-*`. Existing public Pi tool/worker peers cover advertised fields, validation/route refusal, large intervals, both request layers, explicit/disabled/derived deadlines, cancellation/EOF and uncertain cleanup. The final configured check/package results are recorded below. |
+| Windows prerequisites and compilation | Private Windows Node 22.23.4 resolved dependencies from the exact WSL UNC integration checkout; no symlink staging. The final native timing/input helper and disposable controls fixture compiled using the installed Framework compiler. Compilation alone did not count as physical input acceptance. |
+| Text pacing | Registered tool definitions under Windows Node entered 21 dispatched UTF-16 units with Chinese, an emoji, normalized CRLF and a literal tab. Independent owned-editor reads matched at default 50, explicit0 and 25 ms. In the c1-scoped run the observed call times were 1255.6,70.5 and661.7 ms respectively. These are single-fixture observations including call overhead, not an end-to-end/model benchmark or universal lower-delay correctness claim. |
+| Text interruption | A 3000 ms interval with 1000 ms explicit deadline interrupted after one unit, preserved cleanup confirmation and did not replay. Deliberate owned-field focus loss stopped further text. A MAX_SAFE_INTEGER interval with timeout0 was explicitly cancelled after its first unit; settlement and OS held-state observation confirmed release before a fresh operation. |
+| Keyboard | Final `--timing --phase=keyboard` run passed generic modifier aliases; omitted/zero holds; Tab focus changes; timed ordinary/chord keys; foreground-loss release; shorter deadline; and cancelled MAX_SAFE_INTEGER hold with timeout0. Independent fixture messages measured requested 200 ms Tab/F23/Ctrl+F24 holds at 203 ms in this run. The 1000 ms deadline interrupted a 3000 ms F24 hold at 1016 ms. Every observed released state had held0, and final health asserted exclusive control released/no held keys. |
+| Mouse | `--case=timing` passed default click/restoration, zero hold, left/right/middle 200 ms holds, target-loss release, short deadline and cancellation with timeout0. Independent messages measured the timed buttons at 219/204/204 ms; the1000 ms deadline interrupted a 3000 ms hold at 984 ms. OS observation confirmed supported buttons released, cursor/foreground restoration was checked after normal release, and final control handback was asserted. |
+| Reviews | Standards found duplicated browser capability refusal; spec found out-of-scope ignored timing fields on set_value. Both were fixed in 87e1f1b, with public rejection regressions. Both reviews then cleared production changes and the subsequent fixture-only corrections. No remaining findings were reported. |
+
+Final configured verification on the integrated source passed: `npm run check` reported **214 passed, one historical-reference skip** (`/tmp/pi-cu-input-final-check.log`). This includes the corrected phase-selector CLI tests; production and fixture reviews have no remaining findings. `npm run test:package` also passed the local 95-entry archive, scoped metadata, licenses/version, development-file exclusion and documentation links (`/tmp/pi-cu-input-final-package.log`). This was package validation only; the published npm 0.1.0 archive was not replaced.
+
+Live logs are owned-fixture evidence: `/tmp/pi-cu-input-live-text.log` (first partial run at 87e1f1b), `/tmp/pi-cu-input-live-text-rerun.log` (partial run at c1f5ad3), `/tmp/pi-cu-input-live-mouse.log` (successful mouse run, same production code), and `/tmp/pi-cu-input-live-keyboard.log` (successful keyboard-only run at 4b63389). Raw logs are temporary local artifacts; this sanitized receipt records their lasting conclusions.
+
+Review cleanup on 2026-11-10 removed two P3 duplication findings against main: native text/key requests now share one early NUL rejection, and the worker timing tests share a private relay factory while retaining their separate assertions and fresh clients. No input policy or timing default changed. `npm run check` passed 214 tests with one historical-reference skip (`/tmp/pi-cu-input-cleanup-check.log`); the native helper compiled on Windows without desktop input. `npm run test:package` passed the 95-entry archive (`/tmp/pi-cu-input-cleanup-package.log`). Existing scoped desktop evidence was reused; no desktop tests or publication were repeated.
+
+Two fixture failures remain recorded. First, after text cancellation retired the worker, the runner reused an observed element ID; production correctly returned STALE_ELEMENT with no new input. The fixture now observes editors afresh at explicit new-operation boundaries. Second, after an Alt tap, the WinForms message filter missed a subsequent alias key-down in native menu handling, despite confirmed release. The owned timing/text form now suppresses SC_KEYMENU solely for its event oracle; production menu handling is unchanged. The final keyboard phase observed both down/up for Alt and Option. A phase selector allowed rerunning only remaining keyboard cases. No single successful all-phase run or general Alt-menu compatibility is inferred from these composed receipts.
+
+All desktop activity was serialized by the existing shared flock. No clipboard contents, personal documents/profiles, paid model, DSH comparison, display changes or new physical Esc demonstration were used. Existing actual Pi host/stop/display/profile evidence is reused; these live fixtures execute the Pi-registered tool definitions through their established host adapter, not a new real-model or real-Pi-CLI workflow. The acceptance concerns the changed input contract and recorded platform, not every keyboard/device/application or indefinite-duration physical endurance.
+
+## npm publication receipt
+
+On 2026-11-09 the user completed interactive npm publishing authentication and received `+ @ycheung5/pi-computer-use@0.1.0`, with npm stating the package was processing. The submitted archive has 92 entries and SHA-1 `3018e74362be8f5b25739b717d9b455fface1496`, matching the local archive built from source `73616b5`. The user's publication of the prepared package supplies final first-release approval; earlier accepted limitations remain in force.
+
+Initial independent public registry checks returned no 0.1.0 version. An uncached package metadata read confirmed maintainer `ycheung5 <yatfei9@gmail.com>`, with `latest` temporarily pointing to `0.0.0-stage` and 0.1.0 absent. After the processing interval, a fresh registry read confirmed public version 0.1.0 and `latest: 0.1.0`, with maintainer `ycheung5 <yatfei9@gmail.com>`. The public tarball was downloaded independently: 347,809 bytes, SHA-1 `3018e74362be8f5b25739b717d9b455fface1496`, matching the submitted local archive. Registry integrity is `sha512-abMjRQzquBAzEzWGYu9Xac3ZQTw0tpje0hkzBekZs/jGHGk4frmoDjVIcYoZtGCbR6ZlHsI1L73NUxVvKU5m2A==`. [Published package](https://www.npmjs.com/package/@ycheung5/pi-computer-use). The transient staging state resolved without republishing, changing tags or bumping versions. All first-release delivery owners, including #21/#1, are complete with their recorded scopes. This is metadata/download verification, not a fresh registry-installed Windows workflow; existing host/functional receipts and accepted limitations remain unchanged. No runtime changes, desktop/model tests, second publication, GitHub release or Git tag were performed.
 
 ## Final release audit
 
@@ -17,11 +65,11 @@ Audit date: 2026-11-06. Baseline `89bdea3`; this audit reuses the closed capabil
 | Repaired packaging gap | The manifest declared MIT but the root LICENSE was absent. The actual tarball regression first failed (`/tmp/pi-release-license-red.log`); an intermediate link check also rejected the README audit anchor before this section existed (`/tmp/pi-release-license-green.log`). The final check passed after both fixes. This audit adds the standard MIT license and repository/support metadata. No runtime code or license-policy change. |
 | Installation instructions | Built checkout instructions retained. Prebuilt archive extraction + runtime-only dependency install + `pi install .` is documented. Installed Pi 1.4.0 source treats local files as extension resources and local directories as packages; a raw .tgz must be extracted first. Git clones run dependency installation but this package has no prepare build, so raw unbuilt GitHub-source installation remains unsupported. These are source-checked instructions, not a fresh clean-machine pass. |
 | Known limits/failure disposition | Preserve hook-removal limits, explicit Windows activation refusals, retained text pacing, experimental installed Firefox, accepted unverified #28 cases, #6 available-scale scope and #20 recorder/capture-route scope. Earlier failures stay with their repair/collector/environment dispositions; closed owners are not reopened by old wording. |
-| Remaining release gates | #2 is complete on the user-reported second-computer WSL success and explicit acceptance of remaining clean-environment test limitations on 2026-11-09. #21 retains final supported-workflow/limit acceptance and publication reconciliation. The interactive settings editor/reload check was subsequently user-confirmed below. Existing-machine checks do not waive #2. |
+| Remaining release gates | #2 is complete on the user-reported second-computer WSL success and explicit acceptance of remaining clean-environment test limitations on 2026-11-09. The subsequent interactive confirmation, accepted #2 disposition and user-approved npm publication are recorded below. The first-release gates are complete with explicit limits; untested cases are not inferred passes. |
 
 The README's stale #30/cancellation-pending statements and this page's current capability table are reconciled during the audit. Historical dated receipts remain historical. Final packaging/review receipts are recorded here. Responsiveness review reuses prior scoped evidence: warm native reads had roughly 74 ms medians at the dated lifetime-checkpoint candidate, not a current whole-agent benchmark. Explicit SendInput retains the documented 50 ms per-unit gap plus scheduling overhead; no minimum-delay or universal-latency claim is made. No new observed performance defect justifies a fresh benchmark/model programme in this audit. Clean-install provisioning responsiveness remains with #2.
 
-Final local verification passed after the fixes: 187 tests with one historical-reference skip (`/tmp/pi-release-audit-final-check.log`), and a built 91-entry 0.1.0 archive with MIT/license notices, version coherence, development/environment-file exclusion and documentation links verified (`/tmp/pi-release-audit-final-package.log`). Both standards/spec reviews reported no unresolved findings. Root license text follows the already-declared MIT license; runtime code and dependency versions are unchanged. No Windows desktop, physical input, clipboard, personal profile, model call, tag or package publication was used for this audit. The agent-owned audit checkpoint is complete. Subsequent #2 disposition and interactive acceptance are recorded below; #21 final release approval/publication remains open.
+Final local verification passed after the fixes: 187 tests with one historical-reference skip (`/tmp/pi-release-audit-final-check.log`), and a built 91-entry 0.1.0 archive with MIT/license notices, version coherence, development/environment-file exclusion and documentation links verified (`/tmp/pi-release-audit-final-package.log`). Both standards/spec reviews reported no unresolved findings. Root license text follows the already-declared MIT license; runtime code and dependency versions are unchanged. No Windows desktop, physical input, clipboard, personal profile, model call, tag or package publication was used for this audit. The agent-owned audit checkpoint is complete. Subsequent #2 disposition, interactive acceptance and verified npm publication are recorded in this page; #21/#1 are now complete.
 
 Current scope, 2026-11-02: [ADR 0004](adr/0004-pi-first-acceptance.md) makes Pi functionality the release criterion. DSH manual/host/model runs, credentials and parity benchmarks are removed. Dated DSH/paired reports and earlier “remaining comparison” statements below are historical evidence, not current instructions or blockers; the [Pi-first queue](tickets.md#execution-queue) owns unfinished work. Reference-backed scripts are optional historical developer utilities, not required checks. Do not run manual DSH comparisons unless the user explicitly restores that scope.
 
@@ -29,11 +77,11 @@ The next validation cycle follows the [current implementation phases](implementa
 
 ## First-release preparation
 
-Candidate: **`@ycheung5/pi-computer-use@0.1.0`**, not published or tagged. All 25 tools are implemented, with named Windows/WSL2, native/browser, control/handback and agent-workflow receipts in the [capability matrix](#capability-coverage). Installation #2 is complete for its accepted scope and the user confirmed interactive settings/editor/reload. This preparation changes documentation only; prior audit code checks retain their source scope.
+Prepared version: **`@ycheung5/pi-computer-use@0.1.0`**. This section records pre-publication preparation; the subsequent submission/availability state is in the [publication receipt](#npm-publication-receipt). No release tag was created by that preparation. All 25 tools are implemented, with named Windows/WSL2, native/browser, control/handback and agent-workflow receipts in the [capability matrix](#capability-coverage). Installation #2 is complete for its accepted scope and the user confirmed interactive settings/editor/reload. This preparation changes documentation only; prior audit code checks retain their source scope.
 
 The user selected the personal npm scope after discovering the unscoped name belongs to another maintainer. Package/lock names are updated together, public scoped publishing is explicit, and author metadata is `ycheung5 <yatfei9@gmail.com>`. npm sets registry maintainer ownership from the publisher account, independently of the author field. No account email/ownership change or registry publication is performed by this preparation. The scoped tarball name is `ycheung5-pi-computer-use-0.1.0.tgz`; earlier unscoped artifact receipts remain historical. The packaged Pi-host runner now resolves the installed package by its manifest name, including the scope. Scoped-package verification passed: `npm run lint` completed with existing warnings/information, `npm run test:package` built and verified all 91 entries including scoped source/lock/packed-name coherence, author metadata and public access; `npm publish ./ycheung5-pi-computer-use-0.1.0.tgz --dry-run` succeeded for the public registry. No registry publication occurred, and the live Pi-host runner was not rerun for this metadata/path adjustment.
 
-The release README now introduces capabilities, example tasks, setup, control and tested limits, with an original static SVG banner and CI/license/platform badges. The configured artifact check passed with 92 entries, including the banner and resolved local documentation links. SVG XML parsed successfully; no runtime behavior, desktop test or model call changed. Registry publication remains pending account publishing authentication.
+The release README now introduces capabilities, example tasks, setup, control and tested limits, with an original static SVG banner and CI/license/platform badges. The configured artifact check passed with 92 entries, including the banner and resolved local documentation links. SVG XML parsed successfully; no runtime behavior, desktop test or model call changed. The first publish attempt was blocked by account publishing authentication; the subsequent accepted submission is recorded in the [publication receipt](#npm-publication-receipt).
 
 Proposed release notes:
 
@@ -46,7 +94,7 @@ Carry these limits into the public release notes: Windows can silently remove lo
 
 Preparation validation on 2026-11-09: `npm run test:package` passed the 91-entry 0.1.0 archive, license/notices, version, development-file exclusion and documentation links. The sandboxed npm pack child first exited without diagnostics (code 226); the same configured check passed with permitted execution outside that sandbox. No source repair was needed. Prior runtime/portable tests were reused because this preparation changes documentation only.
 
-Final aggregate approval and actual publication stay under #21; #1 closes after publication. No npm registry or GitHub release has been created by this preparation.
+At the preparation checkpoint, final aggregate approval and publication remained under #21 and no registry or GitHub release had been created. The subsequent user-approved npm submission is recorded above; #21/#1 close after registry availability is verified.
 
 ## Second-computer installation report
 
@@ -58,7 +106,7 @@ This report supplies WSL installation/use evidence for #2. A fresh disposable na
 
 On 2026-11-09 the user confirmed all four requested checks passed in their Pi session: `/computer-settings show` displayed settings without errors; `/computer-settings edit` opened and cancelled with Esc; reopening and saving unchanged with Enter succeeded; `/reload` followed by `show` retained the settings without extension errors. This is human-reported acceptance of the terminal settings display/editor/reload path, not an independently captured transcript or a separate tool-result renderer test. The session version and host were not separately collected in this confirmation.
 
-No runtime changes, desktop input, paid-model runs or repeated capability tests were performed by the coding agent for this receipt. Reuse the accepted #35/#36 supported workflows. Final release/known-limit acceptance and publication remain with #21; #2 is subsequently closed with the accepted limitations in the second-computer receipt.
+No runtime changes, desktop input, paid-model runs or repeated capability tests were performed by the coding agent for this receipt. Reuse the accepted #35/#36 supported workflows. The subsequent user-approved publication and remaining registry availability check are recorded above; #2 is subsequently closed with the accepted limitations in the second-computer receipt.
 
 ## Design cleanup checkpoint
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ycheung5/pi-computer-use",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "description": "Native Pi tools for Windows computer use from Windows and WSL2",
   "type": "module",
   "license": "MIT",