Public skills for Claude Code — privacy-scanned and generalized for anyone to use.
Skill Codex
Use when acting as an angel or early-stage investor evaluating a startup opportunity — screening an inbound pitch deck, preparing or running a founder interview, planning or executing due diligence, sanity-checking valuation or SAFE/convertible/term-sheet terms, reading a cap table or modeling dilution, assessing…
Skill Codex
Use when running a client advisory or B2B consulting engagement — preparing a discovery call, writing a proposal or SOW, scoping work, planning research with a source ledger, structuring a problem into issue trees, drafting an executive brief, board deck, or decision memo, QA-ing a client-facing deliverable, reviewing…
Skill Codex
Use when an author is planning, researching, drafting, revising, fact-checking, or quality-assuring a substantive nonfiction book or book-length manuscript, including argument-driven, narrative, prescriptive, business, leadership, self-development, social science, history, culture, explanatory, and hybrid nonfiction.
Skill Codex
Use when researching, architecting, drafting, auditing, or revising a publication-grade long-form article, essay, analytical piece, thought-leadership post, or multi-part editorial series — where thesis, originality, evidence, argument quality, and citation integrity matter more than word count. Also use to audit…
Skill Codex
Use when automating Google Gemini Notebook Enterprise / NotebookLM Enterprise — listing notebooks programmatically, building a notebook/source inventory, adding or removing sources, diagnosing ingestion failures or 401/403/429 errors, or wiring notebook access into an agent (Codex, Claude Code, MCP broker). Also use…
Skill Codex
Use when working with LaTeX/TeX projects — creating, editing, compiling, or debugging .tex/.bib/.bst/.sty/.cls files, fixing PDF build failures and compile-log errors, repairing bibliographies (BibTeX, BibLaTeX, Biber, natbib), formatting equations/tables/figures, converting Markdown or drafts to LaTeX, or packaging a…
Skill Codex
Use when writing, auditing, refactoring, or evaluating any prompt that will be reused — a single copy-ready prompt for a target model (ChatGPT, Claude, Gemini, Perplexity, Llama, Grok, Copilot, image/video models), or a persistent instruction layer for an LLM system (agent/system/developer prompts, orchestrator and…
Skill Codex
Use when writing, editing, or organizing content in an Obsidian vault — notes, frontmatter/properties, wikilinks, callouts, Dataview/Tasks queries, Bases (.base) dashboards, Canvas (.canvas) files, Templater automation, citation/literature workflows, or vault git commits. Also use when setting up an anti-drift vault…
Skill Codex
Use when writing or editing PowerShell commands or scripts, especially when variable names, shell quoting, heredocs, or path/string interpolation may fail, or when writing command blocks for the user to run manually on Windows.
Skill Codex
Use when cloning/forking repos, syncing feature branches with trunk, resolving merge conflicts, cleaning stale branches, setting up CI gates or pre-commit hooks, triaging CI failures, preparing changelogs/release notes/SemVer bumps, or making a repo "AI-agent ready" — safely acquire, inspect, sync, validate, and…
Skill Codex
Use when giving recommendations, comparing or choosing a tool/library/model/service, estimating costs or pricing, proposing a plan or architecture the user will act on, or stating version/benchmark/"state-of-the-art"/best-option/latest claims that could be outdated. Also use when the user asks "is this grounded?"…
Skill Codex
Use when writing new Rust applications, designing Rust architecture, refactoring Rust code, implementing async/concurrency, integrating databases (SQLx/SeaORM/Diesel), configuring Rust CI/CD pipelines, building with axum/actix-web/Tauri/WASM, or performing Rust code reviews — production-grade Rust engineering…
Skill Codex
Use when advising founders, startup teams, accelerators, or early-stage operators — evaluating a startup idea, validating a problem, reviewing a business model, scoping an MVP, planning go-to-market, critiquing a pitch deck, preparing to fundraise, auditing unit economics, or assessing readiness to scale. Triggers on…
Skill Codex
Use when building Tauri desktop or mobile apps, configuring tauri.conf.json, writing Tauri commands/plugins, setting up Tauri permissions/capabilities, debugging Webview issues, optimizing Tauri binary size, configuring Tauri auto-updates, or reviewing Tauri code — production-grade Tauri v2 engineering guidelines…
Skill Codex
Use when asked to test a backend or API, contract-test, security-test an API, load/performance-test, chaos/resilience-test, or verify server correctness — generic, stack-agnostic back-end / API testing process for any service: unit + integration (Testing Trophy, real deps via Testcontainers), API contract…
Skill Codex
Use when asked to UI-test, check accessibility/a11y, run visual-regression, responsive-test, audit the look and feel, verify components across themes/breakpoints/locales, or check localization/RTL rendering — generic, stack-agnostic UI testing process for any web project: visual regression, responsive + cross-browser…
Skill Codex
Use when asked to UX-test, usability-test, run a UX audit, evaluate user experience, check novice-friendliness, assess whether users can complete a flow, or review content/localization clarity — generic, stack-agnostic UX testing process for any product: expert heuristic evaluation (Nielsen's 10), task-based cognitive…
Skill Codex
Use when designing user interfaces, conducting UX research, building design systems, reviewing accessibility compliance, planning usability testing, defining product metrics, writing JTBD statements, structuring information architecture, auditing contrast/target sizes, or evaluating design critiques — UX engineering…
Skill Codex
Use when writing, reviewing, or testing Arabic or bilingual Arabic/English content — UI strings, i18n locales, RTL layouts, store listings, posts, or docs — or when the user reports English leaking into the Arabic version or literal-sounding translations.
Skill Codex
Use when the user mentions a "canary", "context canary", or "canary check", asks to detect context rot / compaction / drift, says "you stopped using my name" or "did you lose context", asks "how degraded is your context", or wants an early-warning system for long agent sessions — installs a per-turn canary signal…
Skill Codex
Use when presenting UI, website, page, or visual design work — new designs, redesigns, mockups, design iterations, or reviewing another agent's design output before it reaches the user — especially after a previous design was rejected as generic or template-like. This is a pre-presentation deny-gate layered on top of…
Skill Codex
Use when handing off work produced in this session to another AI agent — a Codex implementation brief, an external plan review by ChatGPT/GPT Pro or Gemini — when closing a slice/phase/decision that Codex will implement, or when acting as auditor/planner while another agent implements. For crafting or optimizing a…
Skill Codex
Use when planning any build so the magic gets specced as concretely as the plumbing, before calling any build "done", when a solution works but feels like a demo, or when the user says "last 20%", "finish this", "make it actually good", "is this actually done", or "why does this feel flat" — finds and finishes the…
Skill Codex
Use when the user asks the agent to design, review, refactor, secure, optimize, or implement production-grade software, especially architecture reviews, React/Next.js apps, APIs, database migrations, CI/CD or deployment hardening, security review, performance optimization, legacy refactors, and AI-assisted engineering…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: