physics-student-focus

The Physics Student Focus skill is a specialized teaching workflow component designed to assist educators in managing granular student progress. By bridging manual input with automated processing, it allows teachers to maintain a high-resolution view of student performance without relying on broad, generic analytics. This skill is intended for high school physics instructors who need to track conceptual mastery, diagnose specific difficulties during one-on-one sessions, and maintain a historical record of student development.

• Performs automated optical character recognition (OCR) on student answer sheets or handwritten notes using the DeepSeek-OCR integration via SiliconFlow.

• Extracts and summarizes key performance indicators, specifically focusing on weak and strong Knowledge Points (KP).

• Seamlessly updates individual student profiles stored in the backend repository using standardized profiling scripts, ensuring data consistency across the teacher agent ecosystem.

• Supports the integration of qualitative teacher notes, homework discussion logs, and recent assignment data to provide a holistic view of the student's learning trajectory.

• Facilitates the generation of brief, actionable summaries through memory-side effect integration, aiding in long-term pedagogical planning.

• Input requirements include a unique student ID, the context of the assessment (e.g., specific lessons or exam units), and optional evidentiary files like image or PDF answer sheets.

• The tool expects structured input for knowledge points and discussion notes to ensure the generated profile updates are precise and contextually aware.

• Output is strictly derived, high-level profile information; the skill does not store raw, unprocessed scores, focusing instead on competency mapping.

• Operates within the established teacher agent runtime contract, ensuring that every workflow execution is traceable and aligns with pedagogical goals.

• For best results, utilize the CLI quick start format to pass comprehensive discussion data, allowing the system to refine its diagnosis based on both quantitative OCR output and qualitative instructional feedback.