MMCPMarkets
TT

tushaarmehtaa/tushar-skills

Developer tools
12 stars Quality 70 Trend 33MIT
Repository Install

Reusable agent workflows from real projects. Browse, inspect, and install skills for Codex, Claude Code, and Cursor.

Overview

Reusable SKILL.md workflows from real projects. Install them in Codex, Claude Code, or Cursor; see what each skill needs before installing. The directory contains 35 focused skills. Each one owns a meaningful outcome rather than a single code snippet. Each directory is a self-contained Agent Skills package: a standard SKILL.md plus any references needed to run the workflow. The package format is shared, while installation, invocation, and available tools still differ by runtime. Skill pages render the complete package, including every bundled Markdown reference. Links from SKILL.md into specialist guidance stay on the corresponding file and heading, so the directory view is not a shortened preview of the installed skill. The public catalog shows required access and supported surfaces. QUALITY.md defines the interaction, evidence, execution, and evaluation standard every package is expected to meet.

README

slashskills

workflows saved as Agent Skills.

Reusable SKILL.md workflows from real projects. Install them in Codex, Claude Code, or Cursor; see what each skill needs before installing.

The directory contains 35 focused skills. Each one owns a meaningful outcome rather than a single code snippet.

Browse the directory →

Agent Skills, not runtime-specific prompts

Each directory is a self-contained Agent Skills package: a standard SKILL.md plus any references needed to run the workflow. The package format is shared, while installation, invocation, and available tools still differ by runtime.

Skill pages render the complete package, including every bundled Markdown reference. Links from SKILL.md into specialist guidance stay on the corresponding file and heading, so the directory view is not a shortened preview of the installed skill.

The public catalog shows required access and supported surfaces.

QUALITY.md defines the interaction, evidence, execution, and evaluation standard every package is expected to meet.

Developer tools

Install

Select a package, runtime, and scope in the directory installer. For a deterministic global install, target the runtime explicitly:

# Codex
npx skills add tushaarmehtaa/tushar-skills --skill  -g -a codex -y

# Claude Code
npx skills add tushaarmehtaa/tushar-skills --skill  -g -a claude-code -y

# Cursor
npx skills add tushaarmehtaa/tushar-skills --skill  -g -a cursor -y

Omit --skill to install every skill. See the runtime guides for project-scoped installs, invocation, updates, removal, reload behavior, and known limitations:

Claude Code can also install them as plugins. slashskills holds the skills tested in Claude Code. slashskills-labs holds the rest.

/plugin marketplace add tushaarmehtaa/tushar-skills
/plugin install slashskills@slashskills
/plugin install slashskills-labs@slashskills

Plugin skills run with the plugin name as a prefix, such as /slashskills:deploy-check. Details in the Claude Code plugin guide.

Claude app

Only chat-capable skills expose Claude app upload instructions. Skills that need a repository, shell, or local browser show local coding agent required instead. Their ZIP files remain available for inspection, but uploading one does not provide execution equivalent to a local coding agent.

See the Claude app guide for the supported upload path.

Skill catalog

This table is generated from the same catalog and frontmatter used by the site. Do not edit the generated block by hand; run npm run catalog:sync after changing catalog metadata or a description.

Skill Category What it does and when to use it
agent-instructions workflow Create or reconcile scoped AGENTS.md, CLAUDE.md, and Cursor rules from repository evidence. Use when coding-agent guidance is missing, stale, duplicated, or conflicting.
ai-cost-audit ai Audit AI usage, billing, unit economics, routing, caching, batching, and optimization. Use when AI cost, margin, pricing, model choice, or savings must be measured or verified.
ai-product-development ai Design, implement, or audit production AI features with model selection, streaming UX, tool use, safety, evaluation, observability, and cost controls. Use when building AI into an app.
analytics analytics Plan, implement, or audit analytics, error monitoring, health checks, dashboards, and reporting. Use when designing events, consent-aware tracking, incident visibility, or observability.
auth-implementation auth Implement authentication, authorization, tenant isolation, sessions, and protected routes. Use when adding sign-in, identity sync, provider integration, account linking, roles, or access control.
changelog workflow Turn a verified git range into audience-appropriate changelog copy and optional visual assets. Use when preparing weekly updates, release summaries, or social changelog cards.
cold-outreach marketing Research, write, audit, and improve cold messages, introductions, replies, and sequences. Use when contacting prospects, investors, partners, candidates, or other professional recipients.
credit-metering monetization Implement or audit usage credits with atomic reservations, spending, grants, purchases, expiration, refunds, limits, UI state, and an immutable ledger. Use when an app meters consumable usage.
decision-doc planning Facilitate, research, write, or audit decision records with options, evidence, tradeoffs, ownership, and review triggers. Use when a person or team must choose or document an approach.
demo-video workflow Plan, script, build, and verify Remotion product demos, walkthroughs, launch clips, and explainers. Use when a product or technical claim needs a storyboard or rendered video.
deploy-check devops Run a production preflight across release scope, tests, builds, secrets, migrations, dependencies, configuration, observability, and rollback. Use when deploying, pushing, or approving a release.
email-with-resend infrastructure Implement or audit Resend email with templates, queues, preferences, audiences, campaigns, webhooks, and delivery safety. Use when an app needs consent-aware email or Resend repair.
fundraising planning Assess venture fit, pitches, investors, round plans, meetings, and diligence. Use when founders are deciding whether to raise, preparing a raise, or improving an active fundraising process.
humanize workflow Edit or audit prose for generic AI patterns while preserving voice, meaning, uncertainty, and format. Use when writing sounds synthetic, repetitive, over-polished, or unlike its author.
image-editing design Edit supplied images while preserving product identity, exact copy, composition, and previously approved changes. Use when a supplied image needs reference-based edits or successive revisions.
interface-design design Design and implement production web interfaces from product, content, brand, and interaction evidence. Use when creating or redesigning pages, dashboards, workflows, or components.
landing-copy marketing Research, audit, or write evidence-backed landing-page messaging, proof, objections, and calls to action. Use when product, pricing, campaign, or waitlist pages need clearer positioning.
landing-page marketing Design, build, audit, or repair evidence-led landing pages by coordinating message, proof, structure, interface, and verification. Use when a marketing page must perform a clear job.
mobile-first design Audit and repair responsive interfaces from measured narrow-screen evidence without changing product intent. Use when a page clips, overflows, stacks poorly, or fails on touch.
payments-with-dodo payments Implement or repair Dodo Payments checkout, subscriptions, webhooks, entitlements, billing UI, refunds, and verification. Use when adding Dodo billing or auditing its full lifecycle.
performance-diagnosis devops Diagnose web-app performance across build, server, browser, database, and network paths. Use when investigating slowness, resource spikes, reload loops, regressions, bloat, or poor Web Vitals.
product-experiments analytics Design, implement, validate, analyze, and conclude experiments or safe rollouts. Use when testing product behavior with flags, exposure tracking, metrics, staged releases, or A/B tests.
product-launch marketing Plan, implement, audit, or review launches with positioning, assets, distribution, conversion, measurement, waitlists, and follow-up. Use when preparing or learning from a product launch.
product-spec planning Create or audit product briefs and buildable specs covering scope, flows, data, permissions, rollout, and acceptance. Use when planning a new product or a change before implementation.
product-teardown workflow Produce evidence-backed product or competitor teardowns and transferable principles. Use when analyzing positioning, copy, UX, onboarding, pricing, trust, design, or visible retention loops.
rate-limit devops Implement or audit Upstash Redis rate limits with trusted user, tenant, API-key, or IP identifiers, 429 headers, and outage behavior. Use when protecting Next.js routes from abuse or cost.
readme workflow Audit, write, and verify README files for adoption, operation, contribution, or internal orientation. Use when repository documentation is missing, inaccurate, or hard to follow.
remove-ai-slop workflow Audit and remove AI-like design and copy defaults using rendered evidence and confidence scoring. Use when reviewing an interface for generic or repetitive patterns.
search-ready seo Audit, implement, or verify technical SEO, indexation, content, schema, social previews, and AI discovery. Use when preparing or maintaining a live site for search and answer engines.
skill-creator meta Create, improve, merge, and validate portable Agent Skill packages with adaptive workflows and behavioral evals. Use when authoring or reconstructing SKILL.md packages.
social-sharing marketing Audit, implement, and verify canonical URLs, social metadata, preview images, and share links. Use when shared routes are missing, stale, generic, private, or incorrect.
supabase devops Set up, extend, or audit Supabase schema, grants, RLS, migrations, typed clients, storage, and external auth. Use when integrating Supabase or repairing data access and tenant isolation.
ui-copy workflow Audit, write, and implement interface language across actions, forms, states, errors, progress, and notifications. Use when product copy must match system behavior and voice.
user-insights analytics Analyze behavioral segments, retention, activation, funnels, adoption, churn, and monetization. Use when product, event, billing, or qualitative data must inform a product decision.
who-to-sell-to marketing Find your ideal customer profile from real paid and lost deals, with a report of triggers, disqualifiers and a 90-day plan. Use when deciding who to sell to, or why deals stall or don’t renew.

Contributing

Contributions should preserve the portable Agent Skills package contract. Start with CONTRIBUTING.md, then run:

npm run catalog:sync
npm run check
npm --prefix site test --if-present
npm --prefix site run build

Validation covers standard frontmatter, naming, licensing, description portability, line limits, safe relative links and heading fragments, bundled-reference structure and reachability, behavioral eval coverage, catalog coverage, generated ZIP contents, and this README catalog.

V1 scope

V1 supports local Agent Skills in Codex, Claude Code, and Cursor, plus a capability-gated Claude app path for chat workflows. Copilot, Gemini, and OpenAI plugin packaging are outside v1.

License

MIT © 2026 Tushar Mehta.

View the full README on GitHub

Recommended Tools

Try a different keyword or remove a filter.

Install

npx skillfish add tushaarmehtaa/tushar-skills
New to MCP? Start with the mcpskill.org guide →