Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
Skill Claude CodeCodex needs its repo
A toolkit and workflow for creating presentations, interactive prototypes, animated videos, and editable PowerPoint files. It covers formats for talks, reports, prototypes, architecture reviews, and competitive research.
Skill Claude CodeCodex
Interview a vague problem into a buildable spec. Use when the plan is fuzzy and needs stress-testing before code.
Skill Claude Code
A guided workflow for reviewing recent code changes, fixing confirmed problems, and preparing Git commits. Git is the system commonly used to track code changes and share them with a team.
Skill Claude Code
Offload a Claude Code task to a remote Companion sandbox via SSH. Use when the user wants to continue a task in the cloud, run something remotely, or offload work to their Companion sandbox. Triggers on "offload", "run in cloud", "companion run", "remote task", or "send to sandbox".
Skill Codex
Analyze Ascend NPU collective communication profiling data with a DB-first workflow. Use when the user provides clusteranalysisoutput/clusteranalysis.db, rank-level analysis.db, rank-level ascendpytorchprofiler{rankid}.db, together with profilerinfo.json, and asks about HCCL or hcom communication cost, collective…
Skill Claude CodeCodex
Generate high-quality screenshots of Twitter/X posts using the TwitterShots API. Use when the user wants to: capture a tweet as an image, screenshot a tweet, generate tweet image, convert tweet to PNG/SVG/HTML, create tweet screenshot for social media (Instagram, TikTok), or mentions "TwitterShots", "tweet…
Skill Claude CodeCodex
verify from gsmlg-opt/claude-code.
bluesaurel/antigravity-cat-toolkit
Skill Claude CodeCodex
Protects data integrity for new Antigravity users and enables automatic recovery using conversation logs.
Skill Codex
Orchestrates the RLM repo workflow end-to-end with phase gates, locked artifacts, addenda, traceability, and automatic bootstrap/upsert of AGENTS/PLANS scaffolding. Trigger phrases: "Implement requirement ", "Run RLM Phase ", "resume requirement", "lock Phase ", "verify locks".
Skill Claude CodeCodex
A full-project bug audit that has two AI reviewers inspect source files independently and compare their findings.
Skill Claude Code
Generate base authentication feature scaffold with Clean Architecture. Creates shared auth infrastructure (AuthRepository, AuthNotifier, UserProfile, LoginScreen) that /social-login and /phone-auth extend. Run after /core, before specific auth methods.
DonggangChen/antigravity-agentic-skills
Skill Claude CodeCodex
Guide for conducting comprehensive accessibility audits of code to identify WCAG compliance issues and barriers to inclusive design. This skill should be used when reviewing accessibility, ARIA implementation, keyboard navigation, or screen reader compatibility.
Skill Claude CodeCodex
Quick read-only health check — session context, workspace mounts, tool availability, and task snapshot. Use when the user asks for system status or runs /status.
Skill Claude Code
Run one Ralph-style autonomous iteration on GitHub issues / plan files inside the current Claude Code session. In-session port of ralph/afk.sh — drops Docker, claude --print, codex exec, auth tokens, and the MAXITER outer loop, so there is no per-token API cost. Args mirror the original: --issue N (repeatable), --file…
Skill Claude CodeCodex
Deep-dive research on any topic: trends, technologies, concepts, markets, competitors, frameworks, or any subject requiring thorough multi-source investigation. Use this skill whenever the user wants to understand a topic in depth — not make a decision (use 1-5-X or 1-3-1 for that). Triggers on: "research this", "look…
Skill Claude CodeCodex
Use when creating Android UI automation workflows, recording tap sequences, or building replayable device scripts. Triggers on tasks involving Android device interaction, accessibility selectors, or workflow-script.json files. Do NOT use for pure canvas games, OpenGL rendering, or custom drawing surfaces without…
kleberbernardo/claude-code-skills
Skill Claude Code
Auditoria e correção de segurança nível enterprise — AppSec, Cloud Security, DevSecOps, Infra Security. Detecta e corrige riscos do básico ao avançado, incluindo vibe coding e código gerado por IA.
Skill Claude Code needs its repo
Performs a fix-first code review with conditional specialists and a GO/NO-GO verdict.
Skill Claude CodeCodex
Drive the real app end to end, fix what breaks, re-verify. Trigger 'e2e it', 'test it for real'.
Skill Claude CodeCodex
Harden ALB-backed services (LiteLLM, generic APIs) behind CloudFront + WAF. Creates CloudFront distribution with secret origin header, WAF Web ACL with path whitelist, and locks down ALB Security Group to CloudFront-only. Use when deploying any internet-facing ALB service that should not be directly accessible.
Skill Claude CodeCodex
Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 → 4.6, 4.6 → 4.7, retired-model replacements). TRIGGER when: code imports anthropic/@anthropic-ai/sdk; user asks for…
Skill Claude CodeCodex
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright…
SebastianLuser/claude-code-arcane
Skill Claude Code
Android native development with Kotlin + Jetpack Compose: Material Design 3, project config, accessibility, build troubleshooting.
Skill Claude CodeCodex
Triage and fix review findings that already exist on a GitHub PR, then reply and resolve the threads. Use on a CodeRabbit review URL, a PR whose review comments still need working through, or a local /review-pr findings file. When the PR has no findings yet, /review-pr produces them first.
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: