Add Video Learning Agent for short video platforms

Features:
- VideoLearningAgent for automated video watching on Douyin/Kuaishou/TikTok
- Web dashboard UI for video learning sessions
- Real-time progress tracking with screenshot capture
- App detection using get_current_app() for accurate recording
- Session management with pause/resume/stop controls

Technical improvements:
- Simplified video detection logic using direct app detection
- Full base64 hash for sensitive screenshot change detection
- Immediate stop when target video count is reached
- Fixed circular import issues with ModelConfig

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
let5sne.win10
2026-01-09 22:54:57 +08:00
parent 3552df23d6
commit 5b3f214e20
15 changed files with 2317 additions and 1 deletions

View File

@@ -39,6 +39,13 @@ class DashboardConfig:
MODEL_BASE_URL: str = os.getenv("PHONE_AGENT_BASE_URL", "http://localhost:8000/v1")
MODEL_NAME: str = os.getenv("PHONE_AGENT_MODEL", "autoglm-phone-9b")
MODEL_API_KEY: str = os.getenv("PHONE_AGENT_API_KEY", "EMPTY")
MAX_TOKENS: int = int(os.getenv("PHONE_AGENT_MAX_TOKENS", "3000"))
TEMPERATURE: float = float(os.getenv("PHONE_AGENT_TEMPERATURE", "0.0"))
TOP_P: float = float(os.getenv("PHONE_AGENT_TOP_P", "0.85"))
FREQUENCY_PENALTY: float = float(os.getenv("PHONE_AGENT_FREQUENCY_PENALTY", "0.2"))
# Video learning settings
VIDEO_LEARNING_OUTPUT_DIR: str = os.getenv("VIDEO_LEARNING_OUTPUT_DIR", "./video_learning_data")
# Task history
MAX_TASK_HISTORY: int = int(os.getenv("MAX_TASK_HISTORY", "100"))