Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
- agent-skill
- video
- promo
- marketing
Anthropic's official K-12 teaching skills — lesson planning, differentiation, prep and formative checks, with an eval framework.
Source and evaluation framework for the agent skills shipped in Claude for Teachers: k12-lesson-plan-creation (standards-aligned lesson plans), k12-lesson-differentiation (tiered below/at/above versions plus student-needs adaptations with consistent core content), k12-lesson-prep (a prep partner that works through the key student task and leaves a teacher-only note), and k12-check-for-understanding (1–3 item math checks with distractors drawn from documented misconceptions). Co-developed with Learning Commons.
For teachers (or the developers of teacher tooling) who want ground-truth examples of task-shaped skills: concrete instructions, external knowledge grounding, and measurable outputs. Also a rare public look at how Anthropic evals non-coding skills.
In Claude for Teachers: already installed. Elsewhere, load the plugin manually in Claude Code: git clone https://github.com/anthropics/k12-teacher-skills claude plugin marketplace add ./k12-teacher-skills claude plugin install k12-education@k12-teacher-skills
The skills are written assuming the Learning Commons Knowledge Graph connector (included with Claude for Teachers); the README tells you to adapt them if your environment lacks it. US-standards-centric — other curricula need re-grounding.
Repository
anthropics/k12-teacher-skillsCharted
Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
An agent skill that geolocates a photo with nothing readable in it — OSM geometry, elevation skylines and satellite tiles turn “where was this taken?” into a measured answer with an error radius.
A skill that makes AI coding agents map the architecture and declare the blast radius before they touch a line of code — rendered as one standalone HTML page.