Instructional Design Models & Frameworks
ADDIE, ARCS, ASSURE, and UDL address distinct instructional contexts. Literature reviews show that combining complementary frameworks produces stronger alignment than adhering to a single model.
Evidence-based instructional design with peer-reviewed citations.
108 peer-reviewed studies across 11 research domains. Every recommendation includes its evidence tier: T1 meta-analyses and RCTs through T5 expert opinion. Runs in Claude Code and as a native Chrome Extension.
Add to Chrome
Free
or via Claude Code:
git clone https://github.com/savvides/idstack.git && cd idstack && ./setup
Evidence
Every recommendation cites its source.
Stronger evidence tiers take precedence when findings conflict. All 108 references are documented in evidence/references.md with public sources, verifiable citations, and zero proprietary studies.
T1 meta-analyses & RCTs
T2 quasi-experimental w/ controls
T3 systematic reviews
T4 observational
T5 expert opinion & theory
Constructive Alignment & Learning Objectives
Misalignment between objectives, activities, and assessments directly degrades learning retention. Revised Bloom's taxonomy provides measurable action verbs to align evaluation criteria with intended cognitive levels.
Needs Analysis
Evaluates three distinct levels (organizational, task, and learner) before content development begins. Systematic needs assessment prevents courses from addressing peripheral topics instead of core skill gaps.
Cognitive Load Theory & Instructional Sequencing
Working memory limits the volume of novel information learners process simultaneously. Intrinsic and extraneous load principles determine when visual cues clarify concepts and when redundant explanations induce cognitive overload.
Formative Assessment & Feedback
Elaborated feedback that explains error patterns and corrective action yields significantly higher retention than simple correctness indicators. Nicol's principles structure criteria for rubrics, self-assessment, and peer review.
Multimedia Learning Principles
Mayer's cognitive principles (coherence, signaling, segmenting, modality, redundancy) govern effective integration of text and graphics. Violations generate extraneous cognitive load and impair knowledge transfer.
Learner Analysis & Differentiation
Prior knowledge is the strongest predictor of instructional efficacy. Differentiated scaffolding supports novice skill acquisition while avoiding the expertise-reversal effect on advanced learners.
Evaluation Models
Assessment strategies must be designed concurrently with course architecture. Kirkpatrick's four-level model structures evaluation across reaction, learning, behavior, and institutional results.
Rapid Prototyping & Design-Based Research
Formative pilot testing with representative cohorts surfaces usability and conceptual bottlenecks early. Design-based research frameworks ensure iterative revisions maintain methodological rigor.
Online Course Quality Frameworks
Quality Matters (QM) standards and the Community of Inquiry (CoI) model (teaching, social, and cognitive presence) represent the most extensively validated rubrics for online course quality. idstack audits against both.
Accessibility & Universal Design for Learning
WCAG 2.1 AA establishes the legal baseline for digital accessibility. Universal Design for Learning (UDL 3.0) provides proactive pathways (engagement, representation, action) to support diverse learner populations from inception.
Every recommendation includes a research domain code and evidence tier (such as [Alignment-14] [T1]) linking directly to the cited paper. The complete bibliography is available in plain Markdown for verification.
Pipeline
Eleven skills. Pipeline optional.
Eight skills execute sequentially through /idstack:pipeline, while standalone utilities handle course import and institutional memory. You can run any skill independently on existing course files without creating an upfront manifest.
Sequence (left to right) under /idstack:pipeline
- needs-analysis Three-level needs assessment; design-fit gate.
- learning-objectives Bloom's classification; bidirectional alignment.
- assessment-design Rubrics, checkpoints, Nicol's 7 feedback principles.
- course-builder Cognitive-load-aware content generation.
- course-quality-review QM 8 standards + Community of Inquiry.
- accessibility-review WCAG 2.1 AA + UDL 3.0 audit.
- red-team Adversarial audit across 5 dimensions.
- course-export Common Cartridge, SCORM 1.2, or Canvas API.
course-import
Extracts existing modules, assignments, and syllabi from Canvas API, IMSCC, or SCORM.
learn
Manages cross-project institutional knowledge, design preferences, and recurring findings.
Run the complete workflow with /idstack:pipeline. The orchestrator tracks course state, skips completed milestones, and resumes automatically across sessions.
Output
Each skill writes both a report and a manifest.
Every analytical skill outputs two artifacts: a standalone HTML report formatted for faculty and stakeholders, and a structured JSON manifest consumed by downstream skills. Files are organized in .idstack/exports/<course-slug>/.
Finding obj-3: Unassessed Intended Learning Objectives warning T1
Observation. 3 of 8 stated learning objectives have no associated assessment items.
Empirical Evidence. Constructive alignment requires direct correspondence between objectives, activities, and assessments. Unassessed objectives correlate with lower learning gains. [Alignment-14] T1
Recommendation. Run /idstack:assessment-design to generate aligned assessment rubrics, or update the objective list.
{
"learning_objectives": {
"findings": [
{
"id": "obj-3",
"severity": "warning",
"tier": "T1",
"domain": "Alignment-14",
"observation": "3 of 8 ILOs lack assessment",
"consider": "/idstack:assessment-design"
}
],
"report_path": ".idstack/exports/<course-slug>/learning-objectives.html",
"updated": "2026-05-13T14:32:18Z"
}
}HTML reports are self-contained with bundled assets for distribution to faculty and curriculum committees. JSON manifests preserve machine-readable state for downstream tools. The project timeline at .idstack/timeline.jsonl records audit history, and /idstack:pipeline compiles an aggregated index.html dashboard summarizing all findings.
Install
Get started in seconds.
Chrome Extension
In-Browser (Canvas / Docs)Audit Canvas course modules, assignments, Google Docs syllabi, and web course pages directly in a native Side Panel. Zero command-line setup.
Install from Chrome Web Store FreeClaude Code Plugin
CLI / Multi-Agent Pipelinegit clone https://github.com/savvides/idstack.git && cd idstack && ./setup
./setup registers idstack with Claude Code and installs all 11 skills. Re-run it after a git pull to update.
Requires a bash shell (Terminal on macOS or Linux, WSL or Git Bash on Windows). PowerShell and cmd are not supported for setup execution. Python 3 is recommended for quality-score progression and search indexing.
To verify an installation, run bin/idstack-doctor to diagnose plugin registration or bin/idstack-status to view course health.
New to Claude Code? Download it first (free), then paste the command above. When Claude Code starts, run /idstack:pipeline.
What's new
Recent Updates & Release Notes
Chrome Side Panel co-pilot & Course Dossier export.
idstack is officially published as a native Chrome Side Panel extension on the Chrome Web Store. It provides in-browser Bloom's taxonomy classification, constructive alignment reviews, and empirical citations (T1–T5) for Canvas LMS assignments, syllabi, and Google Docs without requiring command-line configuration.
This release also introduces a background Canvas course crawler for multi-page syllabus audits, a Course Dossier export drawer that compiles multi-page findings into structured Markdown (.md) reports, atomic writes for learnings logs (Issue #60), and local-over-global search precedence (Issue #61).
Dedicated Claude Code Plugin Architecture
idstack previously provided an experimental OpenAI Codex CLI build (v2.5.0.0 through v3.3.0.4). The codebase now focuses exclusively on Claude Code to ensure reliable end-to-end testing and performance across all 11 skills.
To update an existing installation, run git pull && ./setup from your local repository. If migrating from the legacy Codex build, see the release notes for manual cleanup steps.
The v3.4.0.1 update calibrated landing page evidence metrics against evidence/references.md, added outbound Canvas API upload and git fetch disclosures to the privacy policy, and clarified Windows shell requirements.
Cross-Platform Python Compatibility & Pipeline Orchestration
Resolved an f-string syntax incompatibility on Python versions prior to 3.12 (including default macOS Python 3.9), restoring session context recovery, score progression tracking, and next-step recommendations. Continuous integration now verifies Python 3.9 compatibility.
This release also corrected namespacing for sub-skills in /idstack:pipeline, restored prior run detection, improved next-step routing for imported courses, and resolved character escaping for course titles with apostrophes.
Patched in v3.3.0.1 (August 5): skill notifications now reference valid runnable /idstack:<skill> commands.
Patched in v3.3.0.2 (August 6): resolved a type-filtering issue when searching learnings records containing apostrophes.
Next: additional specialist skills (content sequencing, media selection, Kirkpatrick evaluation), and push-to-LMS so improvements land back in Canvas without copy-paste. See the roadmap.