Your agent can do more. Find the skill that makes it happen.
Find your next superpower If you are an agent, refer to our llms.txt for full access.Provides Anthropic Imagine design guidance for interactive widgets rendered with show_widget in Claude conversations. Supports charts, dashboards, diagrams, mockups, explainers, and controls such as sliders and toggles.
Rewrite any prompt, instruction, task description, or spec into a "genie-proof" version — instructions so explicit, literal, and loophole-free that even a maliciously literal genie (or an LLM, contractor, or junior dev) could not misinterpret them. Use this skill whenever the user asks to genie-proof, tighten, harden, de-ambiguate, or "make bulletproof" a prompt or instruction; whenever they compl…
Validates BED and universe contracts, plans Region2Vec or scEmbed runs, and checks model and tokenizer compatibility. Also assesses consensus universes for genomic interval workflows.
Runs mutation testing and necessist, then relates surviving mutants and unnecessary test statements to call graph data. Uses this evidence to triage false positives, find missing coverage, and select fuzzing targets.
Provides a terse communication mode intended to reduce token usage by about 75% while preserving technical accuracy. Offers polite, normal, and extreme levels, activated by brevity requests or /genshijin.
Creates compressed commit messages with subjects of at most 50 characters in Conventional Commits format. Emphasizes the reason for a change and supports both Japanese and English.
Rewrites files such as CLAUDE.md, todos, and settings into compressed caveman-style text while retaining code, URLs, structure, and technical content. Replaces the source with the compressed version and saves a readable copy as FILE.original.md.
Explains when to use genshijin investigator, builder, and reviewer subagents instead of inline work or a standard Explore agent. Uses compressed outputs intended to reduce tool-result context by about 60% and support longer sessions.
Shows a one-time reference card covering all genshijin modes, skills, and commands. Does not change the active mode or persist state.
Formats pull request and code review comments as concise, actionable findings without introductory text. Each line identifies a location, problem, and correction, with Japanese support.
Reads token usage directly from Claude Code session logs and displays estimated savings. A mode-tracker hook supplies the output, so the model does not estimate usage or calculate the figures itself.
Optimizes existing content for citation readiness across ChatGPT, Perplexity, AI Overviews, Gemini, and Claude. Focuses on AI citations rather than structural on-page SEO or drafting new content.
Skills give your agent reusable instructions for a specific job. Pick one, read what it does, and bring it into your workflow.
A name only tells half the story. Search the full description and instructions to find the right fit.
Read the skill, visit its source, and see exactly what you’re adding to your agent.
Copy the install command from a skill page and run it in your project.
npx skillycli add owner/repo --skill name