master-instructional-design

A Claude skill that embodies a 30-year veteran instructional design practitioner — coaching, auditing, and elevating every dimension of your L&D practice across 15 engagement modes and 31 on-demand
reference files.

⚖️ Proprietary Framework & Terms of Use

This repository represents the proprietary instructional design methodology developed by Norman
Arosemena, CPTD.

• For Individuals/Learners: You are welcome to use this skill for personal research, study, and
  non-commercial professional development.
• For Organizations/Consultancies: Commercial use, redistribution, or "white-labeling" of this
  framework or its underlying logic is strictly prohibited without a commercial license. For enterprise
  licensing or consulting inquiries, please contact me directly.

What it does

Activates expert-level guidance across the full L&D spectrum:

• Instructional Design — objectives, storyboards, assessments, scenario design, needs analysis,
  task analysis
• eLearning Authoring — Storyline, Rise, Captivate, Lectora, Camtasia, iSpring; advanced JavaScript/CSS
• Facilitation & Live Learning — ILT/VILT workshop design, facilitation guides, train-the-trainer, VILT platforms
• Learner Experience Design (LXD) — learner journey mapping, empathy mapping, experience ecosystems
• Inclusive & Emotional Design — psychological safety, stereotype threat, DEI content design,
  emotional arc
• Agile/Scrum for L&D — sprint-based ID, backlog writing, SAM + Scrum integration
• Visual Design, UX & UI — visual hierarchy, typography, color, usability heuristics for eLearning
• Adobe Creative Suite — Photoshop, Illustrator, InDesign, After Effects, Premiere, XD workflows
  for L&D
• Generative AI for L&D — prompt engineering, AI-assisted ID workflows, agentic pipelines, responsible AI
• CLO & Learning Leadership — organizational learning strategy, talent analytics, skills taxonomy, 3-year L&D strategy
• Academic ID Theory — graduate program canon, major textbooks (Dick & Carey, Gagne, Smith & Ragan, Reigeluth, van Merrienboer, Bloom, Merrill)
• LMS/LXP Strategy — platform selection, RFP process, learning data architecture, xAPI
• Project Management — project charters, RACI matrices, design documents, QA checklists, stakeholder communications
• Evaluation Planning — Kirkpatrick L1–5, ROI calculation, L3 manager observation, learning
  analytics
• Document & Artifact Generation — on-demand production of complete, populated L&D documents: facilitator guides, job aids, storyboards, audience analysis reports, alignment matrices, SME interview protocols, content audits, and communication plans; generation is consultative — the skill diagnoses before drafting and challenges assumptions during the build

Who it's for

• Instructional designers at all levels — from new practitioners to senior IDs
• L&D managers and program designers building or auditing learning products
• Graduate students in ID, LDT, EdTech, or related programs
• CLOs and learning leaders shaping organizational learning strategy
• Anyone who designs, builds, facilitates, or leads learning experiences

Install

Claude.ai (Projects)

1. Download master-instructional-design.skill from the Releases page
2. Open your Claude Project → Skills → Upload skill
3. The skill activates automatically when relevant topics are mentioned

Claude Code

# 1. Clone the repo
git clone https://github.com/narosemena/master-instructional-design
cd master-instructional-design

# 2. Copy the skill files to your global Claude Code skills directory
cp -r master-instructional-design ~/.claude/skills/

# 3. Copy hooks, agents, and settings to your project (or global) Claude config
#    Run this from inside any project where you want the skill harness active:
mkdir -p .claude/hooks
cp .claude/hooks/reference-router.py .claude/hooks/
cp .claude/hooks/guard-writes.py .claude/hooks/
cp .claude/hooks/session-end.py .claude/hooks/
cp -r .claude/agents .claude/agents
cp .claude/settings.json .claude/settings.json  # merge manually if you have an existing one

The reference router hook (reference-router.py) automatically hints which reference file is most relevant for each prompt — zero token cost, keyword-based. Runs on UserPromptSubmit. The guardrail hook (guard-writes.py) protects SKILL.md and all reference files from accidental overwrites via PreToolUse.

Already have a .claude/settings.json? Manually merge the hooks block from this repo's .claude/settings.json into yours rather than overwriting it.

Optional: MCP integrations — each user provides their own credentials. Copy the template and rename it, then set your environment variables:

cp .mcp.json.example .mcp.json   # gitignored — stays local to your machine

Users who don't need MCP integrations can skip this step entirely — the skill and routing hook work without it.

What's inside

Specialist subagents — isolated agents with focused expertise:

SKILL.md — the core skill file with 15 engagement modes, a 9-dimension audit framework, coaching response patterns, diagnostic questions, evaluation planning framework, and the Artifact & Document Output Protocol — a consultative document generation system that diagnoses before drafting, applies the three design lenses during the build, and closes with the most important design risk the document reveals.

31 reference files loaded on demand — never all at once, preserving context efficiency:

Tested against

10 progressive complexity challenges — from writing a single learning objective (L1) to building a
3-year L&D capability strategy for a 5,000-person organization (L10).

Result: 314/320 (98.1%) across alignment, learner-centeredness, cognitive load, practice/transfer, feedback quality, engagement, visual design, DEI & inclusion, emotional design, theory grounding, and
consulting posture dimensions.

Recent Enhancements

v3.2.0 — Specialist Subagents + Session Continuity

• Needs Analyst subagent (.claude/agents/needs-analyst/) — structured one-question-at-a-time intake: performance gap → root cause (taxonomy signal) → audience → constraints → success definition. Does not recommend solutions during intake. Writes completed project brief to memory.json.
• Evaluation Architect subagent (.claude/agents/eval-architect/) — full Kirkpatrick L1–L5 (Phillips ROI) evaluation architecture tied to a specific performance gap. Flags L3 infrastructure gaps and missing baselines. Reads active project memory to tailor the architecture.
• Session log hook (session-end.py) — appends session summaries to session-log.md (gitignored) on Stop. Silently no-ops when no summary is available.
• Notion MCP — .mcp.json.example now includes the Notion server for syncing project briefs, decisions, and risk logs to a shared workspace. Requires NOTION_API_KEY (free with Notion account).

v3.1.0 — Skill Harness Upgrades

• Cross-session project memory — memory.json (gitignored) persists active project state. The UserPromptSubmit hook injects active projects (updated within 30 days) as pre-loaded context. A ## Memory Protocol section in SKILL.md governs when to read, write, and merge.
• PreToolUse guardrail hook — guard-writes.py intercepts Write/Edit calls targeting SKILL.md or any file in references/ and prompts for confirmation before proceeding.
• Dynamic mode hints — the reference router detects unambiguous classification signals (soft-new, soft-change, hard-new, hard-change, mixed) and injects a one-line design cell hint before routing.
• Perplexity + Google Drive + Notion MCPs — .mcp.json.example template for real-time research and document access. Each user provides their own credentials; .mcp.json is gitignored.

How to use it

Just mention what you're working on. The skill activates automatically based on context. Examples:

* *"Audit these 5 learning objectives for a new manager course"*
* *"Design a blended onboarding program for software engineers"*
* *"Our CSAT dropped 11 points — the sponsor thinks agents need product knowledge training. What do I do?"*
* *"Help me write JavaScript to track variable states in Storyline"*
* *"Build a 3-year L&D strategy for a 5,000-person financial services firm"*

🛠️ Work with Me

If you are looking to elevate your organization's learning ecosystem or require a customized deployment of this AI framework, let's connect.

* **Consulting:** Full-spectrum L&D strategy, project management, and UI/UX for eLearning.
* **Workshops:** Training your ID team on AI-Assisted Instructional Design and Agile L&D.

Contact me via LinkedIn or open a private inquiry via GitHub Issues.

Attribution & Disclaimer

This skill references established frameworks, models, and published works in the instructional design
and learning sciences fields. All frameworks and theories are described in original language and
attributed to their respective authors. Tool and product names are trademarks of their respective
owners.

This project is a personal endeavor created by Norman Arosemena. It is not affiliated with, endorsed by, or representative of any current or past employers.

License

CC BY-NC-ND 4.0 — This work is licensed under a Creative
Commons Attribution-NonCommercial-NoDerivs 4.0 International License.

Feedback & Requests

While this is a proprietary framework, I value community feedback. If you identify a gap, an outdated
reference, or a framework that should be explored for the core edition, please open an Issue with a brief description.