From workflow-skills
Interactive, lesson-by-lesson course content review for the learning-center repository. Reviews LMS lessons and lab assignment.md files against the Learning Center style guides and Google Technical Writing best practices. Presents findings conversationally one lesson at a time, pausing between lessons for the reviewer to absorb results. Read-only: flags issues with severity ratings and suggests fixes but never modifies files or leaves PR comments. Use when the user says "review course", "course review", "review lessons", "review this course", or asks for a content/style review while working in the learning-center repository.
npx claudepluginhub arosenkranz/claude-code-config --plugin workflow-skillsThis skill uses the workspace's default tool permissions.
This skill performs an interactive, lesson-by-lesson style and content review of a Learning Center course. It flags issues with severity ratings and suggests fixes but never modifies files or leaves PR comments.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Designs and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.
Designs, implements, and audits WCAG 2.2 AA accessible UIs for Web (ARIA/HTML5), iOS (SwiftUI traits), and Android (Compose semantics). Audits code for compliance gaps.
This skill performs an interactive, lesson-by-lesson style and content review of a Learning Center course. It flags issues with severity ratings and suggests fixes but never modifies files or leaves PR comments.
The review runs in up to three focused passes per lesson, each with a different reading strategy:
| Pass | Name | Reading strategy |
|---|---|---|
| 1 | Structure & Format | Scanning for rule violations (mechanical, rules-based) |
| 2 | Writing Quality | Close reading at sentence/paragraph level (judgment-based) |
| 3 | Content & Flow | Reading as a learner would (narrative-level, holistic) |
Hard constraints:
{target="_blank"} on links.learning-center repository (typically ~/workspace/learning-center/).docs/style-guide/ in the repo. Consult them for edge cases not covered by the pass reference files.courses/<name>/ directory, infer the course from cwd.gh pr list --head $(git branch --show-current) --json number,url --jq '.[0]' to detect an open PR for the current branch.gh api repos/{owner}/{repo}/pulls/{number}/comments.Scan the course directory for reviewable content:
lms/*.md, sorted by numeric prefix (e.g., 01-intro.md, 02-concepts.md).labs/*/, sorted by numeric prefix (e.g., labs/01-setup/, labs/02-instrument/). Each lab track is one review unit. Review all assignment.md files within the track together.Before reviewing any lessons, present:
If PR review comments were found in Step 2, display them under a ## Prior PR Review Comments heading, grouped by file.
Then prompt for lesson selection:
Ready to start reviewing? Or would you like to select specific lessons?
If the user selects specific lessons, adjust the review queue accordingly.
Then prompt for pass selection:
Which review passes would you like to run?
- Structure & Format — heading hierarchy, formatting rules, accessibility, style guide compliance
- Writing Quality — sentence clarity, active voice, paragraph structure (Google Technical Writing)
- Content & Flow — readthrough as a learner, logical flow, step gaps, pacing
Default: all three. Say "1 and 3", "just 2", or "all" to choose.
After pass selection, read only the reference files for the selected passes:
~/.claude/skills/course-review/references/pass-1-structure-format.md~/.claude/skills/course-review/references/pass-2-writing-quality.md~/.claude/skills/course-review/references/pass-3-content-flow.mdFor edge cases not covered by the pass files, consult the raw style guides in docs/style-guide/.
For each lesson in the queue:
After all lessons are reviewed (or the user says "stop"), present:
Before listing detailed findings for a lesson, output a summary table with a Pass column:
| # | Pass | Line(s) | Severity | Rule |
|---|------|---------|----------|------|
| 1 | 1 | 12 | critical | No H1 headings in LMS |
| 2 | 2 | 23-25 | major | One idea per sentence |
| 3 | 3 | 34-40 | major | Missing context switch |
Then group detailed findings by pass, with a pass header for each:
## Pass 1: Structure & Format
[detailed findings]
## Pass 2: Writing Quality
[detailed findings]
## Pass 3: Content & Flow
[detailed findings]
Use this exact format for each detailed finding:
### Finding [N] - [critical/major/minor]
**File:** `path/to/file.md` | **Line(s):** [line number or range]
**Pass:** [1 / 2 / 3] | **Rule:** [Rule name from the pass reference file]
**Issue:** [What is wrong, in one sentence]
> [Quoted offending text, verbatim from the file]
**Suggested fix:**
> [Corrected text or a description of the fix needed]
Group related findings when the same issue appears multiple times in one lesson and the same pass. For example, if passive voice appears on lines 12, 34, and 67, report it as a single finding with all three line numbers.
After all pass findings for a lesson:
**Lesson summary:** [N] findings (critical: x, major: x, minor: x)
Ready for the next lesson? (say **next**, **skip**, or **stop**)
If a lesson has no findings across all selected passes, say so explicitly:
**Lesson summary:** No findings. This lesson looks great.
Ready for the next lesson? (say **next**, **skip**, or **stop**)
| Severity | Definition |
|---|---|
critical | Incorrect information, broken instructions, missing steps that block a learner, accessibility violations (missing alt text, skipped heading ranks, H1 in LMS). |
major | Style guide violations affecting clarity or professionalism: wrong terminology, passive voice in instructions, heading case violations, missing activity/lab summaries, wrong navigation format. |
minor | Style guide violations that don't impede learning (missing Oxford comma, em dash spacing, minor formatting) and improvement suggestions (sentence clarity, word choice, paragraph structure). |
{target="_blank"} on links.docs/style-guide/.