发现 Skills

OpenAI Codex CLI wrapper — three modes. Code review: independent diff review via codex review with pass/fail gate. Challenge: adversarial mode that tries to break your code. Consult: ask codex anything with session continuity for follow-ups. The "200 IQ autistic developer" second opinion. Use when asked to "codex review", "codex challenge", "ask codex", "second opinion", or "consult codex". (gstack)

Scans Cosmos SDK blockchains for 9 consensus-critical vulnerabilities including non-determinism, incorrect signers, ABCI panics, and rounding errors. Use when auditing Cosmos chains or CosmWasm contracts.

Track product champions for job changes and qualify their new companies against ICP. Takes a CSV of known champions (with LinkedIn URLs), creates a baseline snapshot via Apify enrichment, then detects when champions move to new companies. Scores new companies on a 0-4 ICP fit scale. Outputs a downloadable CSV of movers with qualification verdicts.

Update and maintain project documentation for local code changes using multi-agent workflow with tech-writer agents. Covers docs/, READMEs, JSDoc, and API documentation.

Icon usage conventions and semantic mappings. Use when adding, modifying, or choosing icons.

Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-image; 1K/2K/4K; use --input-image.

Generate structured narrative text visualizations from data using T8 Syntax. Use when users want to create data interpretation reports, summaries, or structured articles with semantic entity annotations. T8 is designed for unstructured data visualization where T stands for Text and 8 represents a byte of 8 bits, symbolizing deep insights beneath the text.

Read and parse CHANGELOG.md for version info and release notes. Use when users ask about the latest version, version history, release notes, or changelog content. Triggers on "latest version", "changelog", "release notes", and "version log".

Attempt to fix a security issue tracked in <tracker> by implementing the change in a public <upstream> PR. Runs the sync-security-issue skill first to reconcile the issue's state, then analyses the discussion to decide whether the issue is easily fixable (clear consensus, small scope, known location). If it is, proposes an implementation plan, waits for explicit user confirmation, writes the change in the user's local <upstream> clone, runs the local checks and tests, opens a PR from the user's fork via `gh pr create --web`, and updates the <tracker> tracking issue with the new PR link and any relevant labels. Public PR content is checked to make sure it does not reveal the CVE, the security nature of the change, or any link back to <tracker>.

Download YouTube videos with customizable quality and format options. Use this skill when the user asks to download, save, or grab YouTube videos. Supports various quality settings (best, 1080p, 720p, 480p, 360p), multiple formats (mp4, webm, mkv), and audio-only downloads as MP3.

Identifies non-deterministic or unreliable tests through static code analysis and test result analysis. Use when Claude needs to find flaky tests, analyze test reliability, or investigate intermittent test failures. Supports Python (pytest, unittest) and Java (JUnit, TestNG) test frameworks. Trigger when users mention "flaky tests", "intermittent failures", "non-deterministic tests", "unreliable tests", or ask to "find flaky tests", "analyze test stability", or "why tests fail randomly".

LLM-first SEO analysis skill with 16 sub-skills, 10 specialist agents, and 89 evidence collection scripts for comprehensive SEO audits

Build autonomous self-evolving AI agents with vision-grounded memory that operate computers through a perceive-reason-act cycle

Expert in using and contributing to the claude-skills library — 268+ AI coding agent skills and plugins for Claude Code, Codex, Gemini CLI, Cursor, and more.

Automate iOS and macOS App Store Connect workflows — TestFlight, builds, submissions, screenshots, and metadata using the asc CLI

Turn Hermes into an autonomous project executor using Kanban boards inside an Obsidian vault with visual rendering and REST API control.

Clean AI refusal responses from Codex CLI, Claude Code, and OpenCode session files, and inject CTF/pentest prompts to reduce refusals.

Use DingTalk Workspace CLI (dws) to manage DingTalk contacts, calendar, todos, attendance, approvals, and more from the command line or AI agent workflows.

```markdown

Open-source harness for generating 3D CAD models from text using AI coding agents with build123d/OpenCascade, exporting STEP/STL/URDF, and previewing in a local CAD Explorer viewer.

Responsible AI governance skill for software development. Enforces a 7-pillar RAI framework (tenant/data isolation, PII protection, citation integrity, confidence scoring, hallucination prevention, bias mitigation, content provenance) during development. Use when building or reviewing code touching LLM pipelines, AI-generated output, data source connectors, user data queries, multi-tenant isolation, disclaimers, or transparency features. Use for auditing RAI compliance before releases. Triggers on "responsible AI", "RAI", "citation", "hallucination", "bias", "PII", "confidence score", "content provenance", "disclaimer", "trust page".

Gemini CLI for one-shot Q&A, summaries, and generation.

Create and manage high-quality custom instruction files for GitHub Copilot. Use when you need to define new project-specific guidelines, workflows, or coding standards in the instructions/ directory.

Apple FoundationModels framework for on-device LLM — text generation, guided generation with @Generable, tool calling, and snapshot streaming in iOS 26+.