joacod
- 14 skills
- 0 followers
- 1 day ago last updated
- ▌ Clawbar · joacod bundleA playful bar for OpenClaw agents. Pick a substance, start a session, order rounds, and apply the returned behavior modifiers until the session ends.
- ▌ Dx First · joacod bundleAudit, improve, or guard developer experience (DX/DevEx) across any software repository or project type. Use this whenever a user asks about repository onboarding, development setup, local development, developer tooling, build/test feedback, task ergonomics, confusing commands, CI/local workflow friction, codebase discoverability, reproducible environments, development workflows, repository usability, or reviewing plans and changes that affect developer workflows, even when they do not use the term DX.
- ▌ Rant Coding · joacodTransform existing coding or agent prompts into shorter, sharper, profanity-heavy pissed-off-engineer rants without changing technical meaning. When invoked as `/rant-coding` followed by a prompt, execute the transformed task immediately by default. Use `/rant-coding --prompt-only` or `/rant-coding --dry-run` when the user wants only the transformed prompt. Also use this whenever the user asks to rant, rantify, shorten, make angry, make aggressive, or remove the bullshit from a coding or agent prompt. Do not trigger merely because a user swears during an ordinary coding conversation. The request must be prompt transformation or an explicit `/rant-coding` invocation.
- ▌ Test Health · joacod bundleAudit a repository's automated testing health and recommend a bounded, high-value next improvement slice. Use when assessing an existing test suite, introducing tests into an untested or legacy project, deciding what to test next, evaluating testing strategy or coverage, choosing between unit, integration, and end-to-end tests, investigating brittle or flaky tests, or determining whether a small architecture change is needed for testability. Detect and prefer the repository's existing stack and conventions. Improve testing incrementally rather than pursuing arbitrary coverage targets.
- ▌ Readme First · joacod bundleCreate, improve, simplify, rewrite, or review a software repository's main user-facing README.md. Use this whenever a user asks for a README from an existing codebase, faster onboarding, clearer README structure or developer experience, a 60-second quick start, or a fix for stale or inaccurate README instructions. Focus on the repository front door, not arbitrary Markdown, architecture or API documentation, contributor guides, branding, or landing pages.
- ▌ Ticket To Pr · joacodExplicit-only workflow for turning one coding ticket into a reviewable GitHub pull request: create a ticket branch, implement the work, make sensible commits, push the branch, and open the pull request with `gh`. Invoke this skill through your agent harness's native skill-command mechanism, passing the ticket as its arguments. Do not use it for ordinary coding, branch, commit, push, or pull-request requests without explicit invocation.
- ▌ Project Specs · joacod bundleTurn a finished brainstorm into a durable product specification in the current repository: desired end state, research catalog, architecture with confidence labels, the full version roadmap, near-term implementation tasks, and an agent handoff. Use when the user wants to spec a project or experiment before coding, prepare pre-v0 docs, create an implementation roadmap from a brainstorm, document a project so another agent can implement it incrementally, finish brainstorming and write the docs, or make the project plan and V0 tasks. Also use after a version, task, or experiment lands to update STATUS, learnings, the roadmap, and the next near-term tasks from what was learned. Also use when exploration has clearly ended and the next step is organizing everything needed before implementation. Do not use for ordinary planning questions, small coding tasks, starting implementation, writing the main README, or general Markdown cleanup. Write or update docs/project-slug/ and stop before building.
- ▌ Markdown Quality · joacod bundleReview, improve, or create repository Markdown across README files, docs, guides, API references, ADRs, changelogs, contributor files, templates, and agent-facing instructions. Use this whenever a user asks to clean up, standardize, format, audit, or improve one or more .md files, Markdown consistency, links, headings, examples, accessibility, or documentation quality across a repository. Infer each document's role, audience, renderer, dialect, and maintenance status before editing; use Markdown features when they improve comprehension without inventing content or causing renderer incompatibility. For a main README-only onboarding task, use readme-first instead; use this skill for non-README Markdown or repo-wide work.
- ▌ React Anti Patterns · joacod bundleIntroduce React anti-patterns and common mistakes into existing React codebases for training, review, or teaching. Use when asked to intentionally degrade React performance or code quality while keeping apps functional, or to generate anti-pattern examples for junior developer education.
- ▌ Secure Node Typescript · joacod bundleWrite secure-by-default Node.js and TypeScript applications following security best practices. Use when: (1) Writing new Node.js/TypeScript code, (2) Creating API endpoints or middleware, (3) Handling user input or form data, (4) Implementing authentication or authorization, (5) Working with secrets or environment variables, (6) Setting up project configurations (tsconfig, eslint), (7) User mentions security concerns, (8) Reviewing code for vulnerabilities, (9) Working with file paths or child processes, (10) Setting up HTTP headers or CORS.
- ▌ Business Idea Validator · joacod bundleEvaluate and pressure-test business, product, SaaS, service, and startup ideas using current evidence, competitive research, decision gates, repair pivots, and a small paid validation test. Use when deciding whether to pursue, test, pivot, park, or drop an idea; comparing ideas; exploring a problem space; assessing customer pain, buyer and workflow clarity, market crowding, founder fit, economics, or defensibility; or updating a verdict after customer evidence. Return a compact decision card, not generic encouragement or a full business plan.
- ▌ Refactor Design Patterns · joacod bundleReview existing code or architecture and decide whether a specific design pattern would improve it without overengineering. Use this whenever the user is asking whether a pattern is justified, comparing nearby patterns like Strategy vs State or Adapter vs Facade vs Proxy, or refactoring code with recurring branching, growing state logic, hard-to-extend construction, or inconsistent wrappers. Do not use this for broad system design unless the core question is pattern fit, refactoring shape, or whether a simpler non-pattern refactor is better.
- ▌ UI Design Token Refactor · joacod bundleImprove, refactor, review, and systematize web UI styling with CSS design tokens, cascade layers, Tailwind theme variables, semantic color/spacing/typography tokens, dark mode, accessibility checks, and maintainable component styling. Use this skill whenever the user asks to improve CSS, redesign a site, clean up stylesheets, migrate hardcoded values to tokens, organize Tailwind themes/classes, review a PR for design-token or CSS architecture issues, create or update a design system, add light/dark themes, or make future UI changes easier for agents.
- ▌ Open Source Project Validator · joacod bundleEvaluate and pressure-test open-source project ideas for real utility, adoption, portfolio credibility, community impact, career leverage, ecosystem influence, maintainability, and future opportunities. Use when deciding whether to build, launch, continue, reposition, park, or drop an open-source library, developer tool, self-hosted product, protocol, framework, or public technical project; comparing project ideas; defining success beyond revenue; or interpreting stars, downloads, contributors, production users, case studies, and inbound opportunities. Return a compact evidence-based decision card, not generic encouragement or a full implementation roadmap.