Run a read-only Kempt health check on the local AI-coding-agent state (/.claude, /.codex) and explain the agent-debt report. Use when the user asks why their coding agent feels slow, confused or inconsistent, mentions "agent debt" / "agent 体检" / "agent health", or wants to know what junk has piled up in their Claude…
End-to-end Luma event launch: research comparable events in the target city, draft copy that doesn't read like AI slop, create a Private draft via the admin API, fix the startat / duration / capacity traps, and polish theme, font, and cover. Use this whenever the user wants to publish, edit, or shape an event for…
Audit and safely harden authorized SaaS web applications across frameworks and hosting providers. Use for OWASP-aligned reviews, runtime browser checks, HTTP headers, XSS/CSRF, access control, secrets, payments, uploads, privacy, abuse controls, supply-chain risk, link injection, or security remediation; do not use…
Quick-reference card for all ponytail modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /ponytail-help, "ponytail help", "what ponytail commands", "how do I use ponytail".
Adversarial specification writer. Takes a brief (from grill-me or user) and produces a structured spec.md with YAML frontmatter, requirements, acceptance criteria, and target files. Git-aware pipeline: each run on its own branch, squash-merge on approval.
Use when the user wants to inspect, search, normalize, resume, hand off, or optionally import local Codex, Claude, or Antigravity CLI conversation history. This skill provides a local viewer, provider-specific parsers, native resume commands, a Codex handoff flow, and an explicit experimental Codex-history import path.
Lead free-form implementation work through bounded delegation: define the exact outcome, launch the smallest useful executor team, integrate the result, and prove completion. Use when the user asks Work to build, fix, implement, refactor, or change something without selecting a focused Work skill.
Reveal consequential blind spots after answering. Use proactively for exploration, research, product thinking, architecture, planning, creative work, and decisions, or when the user asks what they missed, what to ask next, or to challenge an idea. Skip routine execution unless a hidden issue could materially change…
Teach, explain, or reshape an answer so the user genuinely comprehends and can use it — leading with the governing idea, showing the mechanism, using worked examples, and checking understanding rather than just simplifying. Use when the user invokes $understand, says they don't follow, asks to learn or understand…
AGENTS.md scaffolder for context engineering. One-shot generator of a durable, agent-agnostic context tree for any repo: an AGENTS.md router plus a context/ directory split into 5 narrative modules (architecture, design-patterns, glossary, operations, debugging) with append-only L3 data files (.jsonl + .yaml).…
Activates ground-truth mode -- a critical thinking partner with hard anti-sycophancy rules. Use whenever someone shares a plan, idea, decision, interpretation, or work product and might be seeking reassurance instead of honest assessment. Trigger on any of: "what do you think," "does this make sense," "is this a good…
Use SideSight for screenshots, images, error screenshots, UI mockups, architecture diagrams, charts, OCR, visual comparison, and videos when the host coding model is text-only. Route each visual task to the matching npx sidesight CLI command.
Create, edit, review, or standardize Simplified Chinese documents using formal academic and technical writing conventions. Use when Codex needs to draft or revise Chinese Markdown, README files, reports, papers, specifications, product documentation, or other prose; correct Chinese-English spacing, full-width…
Turn a vague software idea—or motivation without a clear idea yet—into a verified, explainable first release, or grow an existing project through scoped, testable slices. Use for substantial builds across web apps, backends/APIs, CLI/automation, data/AI, libraries, mobile, desktop, and mixed systems; especially when a…
Turn any codebase into a deeply interactive, single-page HTML learning experience for vibe coders — people who build with AI but want to understand what they built. Trigger whenever someone says "explain this codebase," "help me understand this project," "make a course from this repo," "teach me how this code works,"…
Convert raw HTML pasted from a resume builder's browser inspector (Tailwind arbitrary-value classes, data-editable-text spans, resume--action buttons, SVG icon cruft) into clean semantic HTML/CSS and render it to a single-page PDF. Trigger when the user pastes an HTML dump containing patterns like data-editable-text…
Reverse-engineer a TikTok post from its link — download it, grab key frames, and produce a deep-dive analysis (hook, narrative, monetization, engagement quality, etc). Use this skill whenever asked to analyze, break down, or reverse-engineer a TikTok video/slideshow or URL.
Use when an LLM is doing non-trivial coding work, including implementing features, fixing bugs, refactoring, designing APIs or module boundaries, writing tests, reviewing code, or preparing to commit. Guides the model toward plain data, top-down API design, simple control flow, SOLID without ceremony, boundary parsing…
Use when a user asks what Codex plugins are available or can be used, wants plugin recommendations for a task, asks to find/search/discover installable plugins, asks about the plugin marketplace, or asks whether a plugin can help with a task.
★not rated 2 3mo agoA56 tokens
originalMIT
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: