By mayank-io
Development workflow commands, skills, and doc-conventions. Code review, quality gates, engineering practices, and documentation templates with workflow enforcement.
Run iterative code review with 3 parallel reviewers (PE, Sr SDE, QA) and diminishing fix thresholds across up to 4 iterations.
Review [MK] comments in the current document and apply changes immediately without asking for permission
Run iterative design document review with 3 parallel reviewers (PE, Sr SDE, Domain Expert) and diminishing fix thresholds across up to 4 iterations.
Merge a local worktree into main, update doc statuses, update project plan, and recommend next work
Run iterative ML design review with 3 parallel reviewers (Principal ML Engineer, Sr Applied Scientist, ML Ops Engineer) and diminishing fix thresholds across up to 4 iterations.
Apply tagged review comments (e.g. [MK], [REV]) in the current document immediately, using per-tag interpretation rules from `~/.mstack/dev/feedback-tags.json`. Use when the user says "apply feedback", "apply MK comments", "apply review notes", "apply my edits", or wants tagged comments incorporated directly into a document without an approval gate.
Run iterative code review with diminishing fix thresholds. Use when the user says "review my code", "iterative review", "code review", "review this iteratively", or after completing a feature implementation.
Run iterative design document review with diminishing fix thresholds. Use when the user says "review my design", "design review", "review this design doc", "review the design iteratively", or after completing a design document.
This skill should be used when the user says "finish worktree", "close out worktree", "merge and close milestone", "finish this worktree", "done with worktree", "wrap up worktree", or when a worktree branch is complete and needs to be merged into main AND related documentation status, project plan, and next-milestone recommendations need updating. Combines local merge (via git-worktree:local-merge), doc status updates, project plan updates, and next-milestone recommendation into one workflow. Prefer this over plain local-merge when doc and plan updates are needed.
Run iterative ML design review with diminishing fix thresholds. Use when reviewing designs for ML systems, evaluation harnesses, automated search systems, scoring functions, data pipelines, or any design where statistical soundness, data leakage, overfitting risks, and metric validity are primary concerns.
Modifies files
Hook triggers on file write and edit operations
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.
Claude Code plugin marketplace for development and continuous-improvement workflows.
/plugin marketplace add mayank-io/mstack
Then enable individual plugins:
/plugin install ccimprove@mstack
/plugin install dev@mstack
/plugin install download@mstack
ccimprove — meta improvementAnalyze your Claude Code usage and turn one-off work into reusable patterns.
| Command | What it does |
|---|---|
/ccimprove:surface-usage-patterns | Scan all sessions in ~/.claude/projects/ and surface candidates for skills, plugins, agents, and CLAUDE.md additions |
/ccimprove:make-repeatable | Analyze the current conversation and recommend whether to codify it as a skill or a plugin, then build it |
/ccimprove:clean-permissions | Prune one-off pastes, redundant entries, and dead MCP refs from ~/.claude/settings.json |
dev — development workflowIterative reviews, doc conventions, and feedback loops for serious engineering work.
| Command | What it does |
|---|---|
/dev:code-review-iterative | Up to 4 review-fix iterations with 3 parallel reviewer agents (PE, Sr SDE, QA), diminishing fix thresholds |
/dev:design-review-iterative | Same shape, applied to design documents (PE, Sr SDE, Domain Expert) |
/dev:ml-design-review-iterative | ML-specific variant (Principal ML Engineer, Sr Applied Scientist, ML Ops Engineer) |
/dev:apply-feedback | Apply tagged comments ([MK], [REV], etc.) in the current document immediately, no questions asked |
/dev:review-feedback | Same input, but plan-then-apply with approval gate |
/dev:setup-doc-conventions | Bootstrap docs/templates/, dashboard, Obsidian config, and CLAUDE.md workflow section |
/dev:finish-work-on-local-worktree | Squash-merge a worktree into main, update doc statuses, refresh the project plan, recommend next work |
/dev:summarize-this-session | Two-section conversation recap: ongoing task + continuous-improvement insights |
Bundled assets:
design, impl-plan, cutover-plan, analysis, report, postmortem, kb, ops-guide, playbook, experiment)dashboard.base + dashboard.md)apply-feedback and review-feedback look for tagged comments (e.g., [MK], [REV]) in the current document. The set of tags and how each one should be interpreted is configured per user at:
~/.mstack/dev/feedback-tags.json
The file is auto-created on first run with a default MK entry. To add, modify, or delete tags, edit this file directly. Changes are picked up on the next skill invocation — no restart required.
Schema:
{
"tags": [
{
"tag": "MK",
"from": "Document author (initials = MK for Mayank)",
"content": "Direct edit instructions, factual corrections, rephrasing requests",
"action": "Apply edits as written; treat as imperative; only push back if ambiguous"
},
{
"tag": "REV",
"from": "External reviewer",
"content": "Suggestions, open questions, alternative phrasings",
"action": "Treat as suggestions; flag for discussion before applying"
}
]
}
Field semantics:
tag — short uppercase prefix that appears bracketed in documents (matched case-insensitively).from / content — context to help Claude calibrate trust and tone for these comments.action — interpretation guidance. Does not override the skill's posture (apply-feedback still applies immediately, review-feedback still gates on approval), but informs how matched comments are treated within that posture.download — pull content into local MarkdownSave what you read so it stays searchable, summarizable, and yours.
| Command | What it does |
|---|---|
/download:youtube-transcript | Transcript + metadata + chapters + speakers from a YouTube URL (persistent Chrome profile) |
/download:x-post | Single tweet, full thread, or X Article — auto-detects threads and downloads images |
/download:notion-public-site | Crawl a public Notion site, save every page as Markdown with wikilinks and embedded images |
/download:scribd-document | Pull every page of a Scribd document as zero-padded .jpg files via embed view |
/download:alphaxiv-paper | Structured AI-generated overview of any arXiv paper from alphaxiv.org |
npx claudepluginhub mayank-io/mstack --plugin devDownload content from various sources (YouTube, X, Notion, Scribd, arXiv) as local Markdown files
Analyze Claude Code usage patterns to continuously improve workflows, skills, and plugins
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.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
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
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.