A free, open-source collection of skill files for instructional designers working with Claude (or any AI assistant that supports custom instructions and skill prompting).
Each file in this collection gives Claude the context, frameworks, vocabulary, and judgment it needs to function as a genuine instructional design thought partner — not just a content generator.
These are not prompts. They are skill files — structured knowledge documents written to teach Claude how instructional designers think, what frameworks they use, and what good work looks like in each domain.
Load one (or several) into your AI assistant's context, and Claude becomes a collaborator who understands your craft; not just a tool you have to explain yourself to every time.
- Instructional designers who want AI that speaks their language
- L&D professionals looking to use AI as a design partner, not just a writing assistant
- New IDs building their practice and looking for a knowledgeable thinking partner
- Teams who want consistent AI-assisted design support across projects
| File | What It Covers |
|---|---|
ai_partnership_in_id_work.md |
Load this first. Foundational AI partner posture—the designer is the designer, AI is the thought partner; when to ask vs. generate; feedback format; what good coaching looks like in practice |
needs_analysis_performance_consulting.md |
Diagnosing performance problems, determining whether training is the right solution, performance consulting frameworks |
needs_analysis_reading_the_room.md |
Reading organizational dynamics, decoding stakeholder communication, the human perception layer of needs analysis, delivering inconvenient findings |
when_not_to_design.md |
Recognizing non-training solutions, the is-it-really-a-training-problem test, job aids vs. courses, writing the no-course recommendation |
assessment_design.md |
Writing valid assessments, Bloom's Taxonomy, question types, alignment to objectives, AI-generated item failure modes |
writing_for_elearning.md |
Voice, tone, plain language, writing for the screen, narration scripts, humor and levity as a design tool |
strategies_for_managing_cognitive_load.md |
Cognitive load theory, working memory, instructional strategies that reduce extraneous load |
id_models_and_methodologies.md |
ADDIE, SAM, Action Mapping, Gagne's Nine Events, Bloom's Taxonomy, Merrill's First Principles, Kirkpatrick-Phillips, 70-20-10 |
| File | What It Covers |
|---|---|
scenario_based_learning.md |
Case-based and branching scenarios, realistic decision-making, consequence design, LLM distractor failure modes |
interaction_design_basics.md |
Interaction types, engagement design, meaningful vs. cosmetic interactivity, naming the experience not the mechanism |
designing_for_motivation.md |
ARCS model, intrinsic/extrinsic motivation, relevance, learner engagement strategies |
designing_for_behavior_change.md |
COM-B, Fogg's Behavior Model, habit formation, closing the knowing-doing gap |
designing_for_skills_development.md |
Deliberate practice, novice-to-expert continuum, transfer design, tacit knowledge |
culturally_situated_scenario_writing.md |
Cultural texture beyond surface diversity, communication styles, hierarchy, regulatory context, cross-cultural scenario design, working with AI on culturally specific content |
| File | What It Covers |
|---|---|
visual_design_principles.md |
CRAP principles, typography, color, layout, hierarchy, imagery for e-learning |
accessibility_in_elearning.md |
WCAG, Universal Design for Learning, situational accessibility, testing, compliance |
quality_assurance_testing_for_elearning.md |
Functional, instructional, editorial, and accessibility QA; bug tracking; sign-off process |
character_design_for_elearning.md |
Visual style spectrum (photorealistic, flat vector, hand-drawn, comic, cartoon); character consistency across renders; inclusive representation including naming, hair, facial features, age, gender/role distribution, and professional clothing diversity; AI generation constraints; the reference image anchor technique |
designing_for_sensitive_topics.md |
Tone calibration for DEI, harassment, mental health, ethics, and crisis content; the empathy-utility balance; levity as a design tool; process guidance for sensitive content projects |
assessing_interaction_quality.md |
Six-dimension rubric for evaluating scenario and interaction design quality: scenario realism, choice architecture, consequence design, cognitive load calibration, objective alignment, and feedback depth; coaching prompts per dimension; common failure patterns; guidance for using the rubric in ID mentorship conversations |
ui_control_design_for_elearning.md |
The additive redundancy failure mode in AI-generated UI; the principle that controls should communicate their own purpose; design guidance for toggles, sliders, buttons, drag zones, and progress indicators; prompting strategies for AI-assisted control design; proportion and scale; quick review checklist |
| File | What It Covers |
|---|---|
authoring_tool_capabilities.md |
Capabilities and constraints of Storyline 360, Rise 360, and Lectora — interaction types, accessibility, review tools, publishing formats, and when to use which tool |
Relationships are not supplementary to instructional design — they are foundational to it. This folder covers the human side of the work: the partnerships, communication, and community that determine whether great design actually reaches learners.
| File | What It Covers |
|---|---|
stakeholder_management_performance_consulting.md |
Managing stakeholders, scope, pushback, and project dynamics |
facilitation_and_cohort_learning.md |
Designing and facilitating group and cohort learning; psychological safety; virtual and hybrid facilitation |
ai_prompt_engineering_for_ids.md |
Writing effective prompts for ID tasks, prompt frameworks, AI-assisted design workflows |
industry_specific_design_considerations.md |
LLM failure modes and design guidance for healthcare, financial services, legal, government, manufacturing/safety, education, and nonprofit verticals; expert review requirements; industry-specific prompting guidance |
⚠️ Two different upload contexts — two different rules:
- Regular chats: Do not upload
.mdfiles as attachments — this produces an error. Paste the raw text instead.- Claude Projects: Uploading
.mdfiles as Project Knowledge works correctly and is the recommended approach.See
USAGE_GUIDE.mdfor full setup instructions.
Copy the contents of any skill file and paste it at the start of your conversation with Claude before making your request.
If your AI tool supports system prompts or custom instructions, paste one or more skill files there for persistent context.
If you're using an AI orchestration tool that supports skill files natively, these files are structured to work as drop-in skills.
Each skill file is structured with clear headings. You can copy a single section — a framework, a checklist, a set of prompt templates — and use it without loading the full file.
Once you've loaded a skill file, try prompts like these:
With needs_analysis_performance_consulting.md:
"I have a stakeholder who wants a course on time management. Help me think through whether training is actually the right solution."
With scenario_based_learning.md:
"Help me write a branching scenario for new managers who need to practice delivering difficult feedback."
With assessment_design.md:
"Review these five quiz questions and tell me whether they're testing application or just recall."
With accessibility_in_elearning.md:
"Audit this course storyboard for accessibility issues I should address before development."
With quality_assurance_testing_for_elearning.md:
"Help me build a QA checklist for a branching scenario course I'm about to hand off to a client."
With ai_partnership_in_id_work.md:
"I'm about to start a new course design project and I want you to function as a genuine thought partner — not just a content generator. Here's the context..."
With needs_analysis_reading_the_room.md:
"I have a discovery meeting with a stakeholder tomorrow who has already decided they want a compliance course. Help me think through what questions to ask and what to listen for beyond what they tell me directly."
With when_not_to_design.md:
"My needs analysis is suggesting the performance gap isn't really a training problem. Help me build the case for a different solution and draft a recommendation I can bring to the stakeholder."
With culturally_situated_scenario_writing.md:
"I'm designing a management training scenario for a financial services team in Singapore. Help me think through the cultural dimensions I need to get right and what I should verify with a cultural reviewer."
With designing_for_sensitive_topics.md:
"I'm building harassment prevention training and I want it to actually prepare people for real situations—not just check a compliance box. Where do I start?"
With assessing_interaction_quality.md:
"Review this branching scenario interaction and tell me where the choice architecture and consequence design are strong—and where they're falling short."
With ui_control_design_for_elearning.md:
"I need a toggle for switching between two modes in my e-learning UI. Help me design a control that signals its own purpose without needing supplementary labels or icons."
With industry_specific_design_considerations.md:
"I'm designing clinical training for nurse practitioners in an acute care setting. What do I need to know about this vertical before I start, and what should I flag for SME review?"
With character_design_for_elearning.md:
"I need to generate a set of characters for a branching scenario. Help me write a character specification that will keep them visually consistent across multiple emotional states and poses — and make sure the cast reflects the diversity of the real workforce."
With designing_for_behavior_change.md:
"My client wants a compliance course that actually changes behavior, not just checks a box. Where do I start?"
Every skill file in this collection was built with the same commitments:
Performance over coverage These files teach Claude to think about what learners will do, not just what they will know. The language of performance — observable behavior, job application, transfer — runs through everything.
Evidence-informed Frameworks and strategies are grounded in learning science and established ID practice — not trends or buzzwords.
Opinionated where it matters Good instructional design requires judgment, not just information. These files give Claude the stances and heuristics to have useful opinions — not just present options.
Practical, not academic These are working tools, not literature reviews. Every section is written for use on actual projects.
Have a skill to add? See a gap? Want to improve an existing file?
Contributions are welcome. Please:
- Follow the existing file structure and formatting conventions
- Use snake_case filenames with
.mdextension - One skill per file
- Include a references section with credible, practitioner-accessible sources
- Write for working IDs, not academics
Open a pull request or start a discussion in Issues.
This collection was created by Trina Rimmer — instructional designer, L&D consultant, and author of the Kickstart Your E-Learning Career guidebook.
Trina helps aspiring instructional designers break into the field and helps organizations build learning experiences that actually work.
- Website: trinarimmer.com
- Guidebook: trinarimmer.com
Built for instructional designers who want AI that understands their craft.