By smileynet
Real-world concepts from programming, education, and design — packaged as AI commands
Find what you don't know you don't know — surface hidden assumptions, failure modes, and missing perspectives
Research and produce a structured best-practices and antipatterns document
Map all possible causes of a problem using fishbone (Ishikawa) categorical decomposition
Explore and map how a codebase or subsystem works
Systematic root cause analysis — trace symptoms to their actual cause before attempting fixes
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Real-world concepts from programming, education, and design — packaged as AI commands.
These aren't prompt engineering techniques I invented. They're established concepts from real fields — idiomatic code (programming), the Socratic method (education), Diataxis + progressive disclosure (documentation theory), task graphs (operations research), systematic debugging (reliability engineering), program comprehension (software engineering). I just got tired of re-explaining them every session and made commands to use them.
Modifier -- you invoke it once, and it changes the whole session:
> /spell:idiomatic Python 3.12
IDIOMATIC MODE: PYTHON 3.12
══════════════════════════════════════════════════════════════
Version: 3.12
Source: https://docs.python.org/3.12/
SESSION CONSTRAINTS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
USE (canonical patterns):
• pathlib.Path over os.path — modern path handling
• match/case for structural pattern matching — added in 3.10
• f-strings over .format() — cleaner, faster
• type hints with built-in generics (list[int] not List[int]) — PEP 585
AVOID (deprecated or non-idiomatic):
• os.path.join() → Use pathlib.Path / operator instead
• typing.List, typing.Dict → Use built-in list[], dict[] generics
• implicit string concatenation → Use f-strings or explicit +
These constraints are active for the rest of this session.
Workflow -- does research and gives you something concrete:
> /spell:bpap Git commit messages
BEST PRACTICES & ANTIPATTERNS: GIT COMMIT MESSAGES
══════════════════════════════════════════════════════════════
Source: Built-in knowledge + https://cbea.ms/git-commit/
DO ✓
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
1. Use imperative mood in subject line — "Add feature" not "Added feature"
2. Keep subject to 50 characters — forces concise summary
3. Separate subject from body with blank line — tools depend on this
4. Wrap body at 72 characters — readable in terminal and git log
DON'T ✗
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
1. Don't end subject with period — it's a title, not a sentence
2. Don't use "fix bug" or "update code" — say what and why
3. Don't put everything in the subject — use the body for detail
ANTIPATTERNS ⚠
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
1. THE KITCHEN SINK — Unrelated changes lumped into one commit
Why it's tempting: You're "almost done" and don't want to split work
Consequences: Impossible to revert one change without losing others
Instead: One logical change per commit, use interactive staging
Modifiers change how the AI behaves for the rest of the session. You invoke one, then keep working. The constraints stick.
| Spell | What it does |
|---|---|
| idiomatic | AI keeps hallucinating APIs that don't exist? Sets session constraints based on real docs and canonical patterns, not guesswork. |
| socratic | Want to actually learn, not just get an answer? Flips the AI into question-mode so you reason through problems yourself. |
Workflows do research and produce actual output. They'll ask you questions first, then go do the work.
npx claudepluginhub smileynet/sams-genai-spells --plugin spellSimple workflow orchestration for Claude Code sessions - mise, prep, cook, serve, tidy, plate, run
Code quality spice for Line Cook — decision frameworks, antipattern detection, refactoring patterns, tradeoff analysis, YAGNI guidance, dead code pruning, and plan auditing
Game design spice for Line Cook — MLP scoping, core loop frameworks, economy design, difficulty curves, content planning, mechanics palette, playtesting, scenario walkthroughs, anti-patterns, architecture auditing, with interactive commands, code review agent, and implementation guidance
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Real-time statusline HUD for Claude Code - context health, tool activity, agent tracking, and todo progress
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer