feat: parallel testing support and screen recording for SDK#16
Merged
Conversation
- Add `deviceUdid` option to pin specific device per AppClaw instance - Add `video` option to record screen and embed mp4 in report - Auto-allocate unique ports (systemPort, mjpegServerPort, wdaLocalPort) per instance to prevent parallel test collisions - Pass appResolver through session connect to step runner - Improve screenshot handling: use vision pre-action shot (tap dot overlay) when available, fallback to post-action in DOM mode - Attach tapCoordinates and deviceScreenSize to step reports - Add test:e2e:android npm script and android e2e test file - Bump version to 0.1.7 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
…K tests Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
…iability - Skip oscillation/repetition stuck signals for data-entry goals (PIN, timer, keypad) where ABAB hash alternation is expected during digit correction - Force vision-locate for keypad digit/backspace selectors via new vision-tap-policy module (LLM normalized coords too inaccurate on dense pads) - Run done-verification unconditionally (not just in vision+screenshot mode) so DOM-only runs also guard against hallucinated completion - Add computePerceptionHash that fingerprints screenshots when DOM is empty, fixing stuck detection in vision-only mode - Infer drag/swipe destination from direction keywords when vision returns zero/null coords for the target element - Give each sub-goal the full MAX_STEPS budget instead of a proportional share - Add Gemini 3.x thinking support (LLM_GEMINI_THINKING_LEVEL, LLM_GEMINI_INCLUDE_THOUGHTS) and expose reasoning-delta chunks in stream - Tighten done/skip prompts across planner, agent loop, and tool description to require specific visible evidence Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: Srinivasan Sekar <srinivasan.sekar1990@gmail.com>
github-actions Bot
pushed a commit
that referenced
this pull request
Apr 16, 2026
## 1.0.0 (2026-04-16) ### Features * integrate ai-sdk-ollama for LLM support and update configuration ([#9](#9)) ([c6794d7](c6794d7)) * parallel testing support and screen recording for SDK ([#16](#16)) ([7d14e7b](7d14e7b)) * strict playground tap matching, waitUntil pre-check, faster vision assert ([59b8c29](59b8c29)) * vision improvements — drag support, screenshot optimization, an… ([#7](#7)) ([8cfbcb4](8cfbcb4)) ### Bug Fixes * add semantic-release for automated versioning and npm publishing ([#19](#19)) ([66c73a6](66c73a6)) * ci ([#10](#10)) ([dfcd62f](dfcd62f)) * DOM locator resolution, vision assert parsing, and appium-mcp coordinate scaling ([9272c36](9272c36)) * read CLI version from package.json instead of hardcoded string ([#14](#14)) ([fcb3a64](fcb3a64)) * screenshot parsing ([e449a23](e449a23)) * scroll-aware stuck detection, press_enter tool, and post-done verification ([c03bbe4](c03bbe4)) * terminal UI ([294a780](294a780)) * terminal view ([42c0e75](42c0e75))
|
🎉 This PR is included in version 1.0.0 🎉 The release is available on: Your semantic-release bot 📦🚀 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
deviceUdidoption to pin specific device per AppClaw instancevideooption to record screen and embed mp4 in report