explicit-instruction-sequence-builder

这段内容是审核时保存的快照。外部来源才是完整且最新的版本。

---
# AGENT SKILLS STANDARD FIELDS (v2)
name: explicit-instruction-sequence-builder
description: "Build a complete explicit instruction sequence from teacher modelling through guided practice to independent work. Use when teaching new skills, procedures, or concepts through direct instruction."
disable-model-invocation: false
user-invocable: true
effort: medium

# EXISTING FIELDS

skill_id: "explicit-instruction/explicit-instruction-sequence-builder"
skill_name: "Explicit Instruction Sequence Builder (I Do / We Do / You Do)"
domain: "explicit-instruction"
version: "1.0"
evidence_strength: "strong"
evidence_sources:
  - "Rosenshine (2012) — Principles of Instruction: research-based strategies that all teachers should know"
  - "Pearson & Gallagher (1983) — The instruction of reading comprehension (gradual release of responsibility model)"
  - "Archer & Hughes (2011) — Explicit Instruction: Effective and Efficient Teaching"
  - "Hattie (2009) — Visible Learning: direct instruction effect size 0.59"
  - "Engelmann & Carnine (1982) — Theory of Instruction: principles and applications"
input_schema:
  required:
    - field: "skill_to_teach"
      type: "string"
      description: "The specific skill or concept to be taught through explicit instruction"
    - field: "student_level"
      type: "string"
      description: "Age/year group and prior knowledge level"
    - field: "lesson_time"
      type: "string"
      description: "Available lesson time in minutes"
  optional:
    - field: "common_misconceptions"
      type: "array"
      description: "Known misconceptions or errors students make with this skill"
    - field: "student_profiles"
      type: "array"
      description: "From context engine: ability range, EAL needs, SEND profiles"
    - field: "prior_knowledge"
      type: "string"
      description: "What students already know that this builds on"
    - field: "success_criteria"
      type: "string"
      description: "From context engine: how success will be measured"
output_schema:
  type: "object"
  fields:
    - field: "i_do"
      type: "object"
      description: "Teacher modelling phase with scripted explanation and think-aloud"
    - field: "we_do"
      type: "object"
      description: "Guided practice phase with structured teacher-student interaction"
    - field: "you_do"
      type: "object"
      description: "Independent practice phase with monitoring checkpoints"
    - field: "cfu_points"
      type: "array"
      description: "Checking for understanding moments embedded throughout"
    - field: "timing_guide"
      type: "string"
      description: "Suggested time allocation across the three phases"
chains_well_with:
  - "checking-for-understanding-protocol-designer"
  - "worked-example-fading-designer"
  - "think-aloud-script-generator"
  - "practice-problem-sequence-designer"
  - "pedagogical-content-knowledge-developer"
teacher_time: "5 minutes"
tags: ["explicit-instruction", "gradual-release", "modelling", "direct-instruction", "scaffolding"]
---

# Explicit Instruction Sequence Builder (I Do / We Do / You Do)

## What This Skill Does

Generates a complete gradual release of responsibility sequence for teaching a specific skill: a scripted "I Do" (teacher models with think-aloud), a structured "We Do" (guided practice with teacher-student interaction), and a designed "You Do" (independent practice with monitoring points). The output includes checking-for-understanding moments at each transition and a timing guide. AI is specifically valuable here because effective explicit instruction requires the teacher to make invisible expert thinking visible — breaking down a skill they perform automatically into discrete, teachable steps with articulated reasoning. This decomposition of expert performance is cognitively demanding and is where most explicit instruction falls short.

## Evidence Foundation

Rosenshine (2012) synthesised decades of research into ten Principles of Instruction, with explicit instru