Skip to content

Release v0.2.0

Latest
Latest

Choose a tag to compare

View all tags
@mmmay0722 mmmay0722 released this 05 Dec 06:14
v0.2.0
540c351
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

🚀 Release v0.2.0

📦 Docker Images

  • mmmay0722/webqa-agent:v0.2.0
  • mmmay0722/webqa-agent:latest

📝 Changes since v0.1.6

  • v0.2.0: Two-phase Planning, GPT-5 Responses API & Enhanced Execution (540c351)
  • Merge remote-tracking branch 'origin/main' into dev_0.2.0 (2fce5ce)
  • refactor: change Scroll action to element-based navigation (0eb668a)
  • checkpoint: minor update (5a47221)
  • checkpoint: fix cookie injection (3626c6c)
  • update webqa to webqa agent in html (f83742c)
  • checkpoint: 1. fix dynamically generate steps and optimize its prompt; 2. optimize verify under comparison mode; (98ae5c0)
  • feat: Migrate GPT-5 to the Responses API (f35ac22)
  • feat: add GPT-5 reasoning/text controls and enhance test planning prompts (f991d6a)
  • checkpoint: support window.history.back() for different domain scenarios. (057c043)
  • feat: Add uv package manager support with improved dependency management (2a322ac)
  • checkpoint: 1. support comparing before state with after state in verify function; 2. add focus_region parameter in verify function; 3. add language detection prompts; 4. unify single-tab mode among all prompts; 5. remove functional_criticality and cleanup_requirements phrase in planning_prompts.py; 6. add instructions for avoiding specific element id in cases (87ae5b6)
  • checkpoint: minor fix (c33d4c8)
  • checkpoint: single-tab support (5fb52fb)
  • checkpoint: support multi-tab feature (25cf0c6)
  • feat: 1. update docs and example config fix: 1. change the error level of uncrawlable scenarios (09bd66c)
  • minor update (b686a5d)
  • minor update (06f6826)
  • feat: 1. add switch for saving screenshots to local; add context for verify stage fix: 1. remove deprecated prompt in action stage; 2. abort steps and cases when unable to crawl the current page (ca2a84b)
  • Enhanced multiple features and fixed issues (#86) (8e126ce)
  • feat: 1. support container scroll; fix: 1. coordinates conversion; 2. clarify screenshot parameters (82b7c0e)
  • fix: update agent prompt example (049ca9a)
  • feat: add UX verification tool and case recording system (#84) (b99e6ae)
  • feat: 1. support max tokens and filter parameters; 2. support interactive elements and text via LLM filtering; 3. support dynamically plan recovery steps (c704a56)
  • [WIP]feat: support full page planning and scroll (3084ddb)
  • feat: 1. add schema for action tool and verify tool; 2.synchronize actions with action_executor.py; 3. update crawl parameters (8322704)
  • feat: 1. update docs and example config fix: 1. change the error level of uncrawlable scenarios (a526f15)
  • minor update (4fd0b0d)
  • minor update (d0d570c)
  • feat: 1. add switch for saving screenshots to local; add context for verify stage fix: 1. remove deprecated prompt in action stage; 2. abort steps and cases when unable to crawl the current page (e41da1e)
  • Enhanced multiple features and fixed issues (#86) (693737a)
  • feat: 1. support container scroll; fix: 1. coordinates conversion; 2. clarify screenshot parameters (05096c6)
  • fix: update agent prompt example (190c3ae)
  • feat: add UX verification tool and case recording system (#84) (72d2bc5)
  • feat: 1. support max tokens and filter parameters; 2. support interactive elements and text via LLM filtering; 3. support dynamically plan recovery steps (2535171)
  • [WIP]feat: support full page planning and scroll (a1a3710)
  • feat: 1. add schema for action tool and verify tool; 2.synchronize actions with action_executor.py; 3. update crawl parameters (86dccf7)
Assets 2
Loading