发现 Skills

```markdown

Penetration testing (OWASP methodology).

World-class ML engineering skill for productionizing ML models, MLOps, and building scalable ML systems. Expertise in PyTorch, TensorFlow, model deployment, feature stores, model monitoring, and ML infrastructure. Includes LLM integration, fine-tuning, RAG systems, and agentic AI. Use when deploying ML models, building ML platforms, implementing MLOps, or integrating LLMs into production systems.

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes

Interpret macroeconomic indicators including GDP, inflation, unemployment, interest rates, and exchange rates to assess economic health and predict trends. Use this skill when the user needs to evaluate a country's economic outlook, understand monetary/fiscal policy impacts, or contextualize business decisions within the macroeconomic environment — even if they say 'is the economy doing well', 'what do rising interest rates mean for us', or 'explain today's economic data'.

Apply structured problem-solving using the MECE principle, issue trees, a hypothesis-driven approach, and the Pyramid Principle. Use this skill when the user faces a complex, ambiguous problem and needs to decompose it systematically, structure a consulting-style analysis, or organize recommendations clearly — even if they say "where do I start", "this problem is too big", "help me break this down", or "structure my thinking."

Patterns for delegating tasks to Codex CLI (GPT 5.2) experts including bug analysis, code review, and plan revision via the mcp__codex__codex tool.

Extract text from PDFs and scanned documents. Use web_extract for remote URLs, pymupdf for local text-based PDFs, marker-pdf for OCR/scanned docs. For DOCX use python-docx, for PPTX see the powerpoint skill.

Convert documents (PDF/text) into beautifully typeset HTML. Keep the original content completely unchanged and only optimize the visual presentation. Suitable for beautifying service proposals, quotes, planning documents, and similar materials. Users can print directly to PDF from the browser.

Capture architectural decisions as structured ADRs during Claude Code sessions. Automatically detects decision moments and records context, alternatives, and rationale.

Web search and content extraction via Brave Search API. Use for searching documentation, facts, or any web content. Lightweight, no browser required.

Fast file-name and content search using `fd` and `rg` (ripgrep).

Transaction-verified trading signals on Base. Register an agent as a signal provider, publish trades with on-chain TX hash proof, and consume signals from top performers via a REST API. All track records are verified against blockchain data — no fake performance claims. Triggers on: "publish signal", "post trade signal", "register provider", "subscribe to signals", "copy trade", "bankr signals", "signal feed", "trading leaderboard", "read signals", "get top traders".

Mine LITCOIN — a proof-of-comprehension and proof-of-research cryptocurrency on Base. Use this SDK when a user wants to mine crypto with AI, earn tokens through reading-comprehension or solving optimization problems, stake LITCOIN, open vaults, mint LITCREDIT, manage mining guilds, deploy autonomous agents, or interact with the LITCOIN DeFi protocol.

Cross-chain token swaps across 54+ blockchains via the Symbiosis protocol. Use when the user wants to swap or bridge tokens between any chains — Base, Ethereum, Polygon, Arbitrum, Optimism, BNB Chain, Avalanche, Solana, Bitcoin, TON, Tron, and 40+ more. Supports any-to-any token swaps with automatic routing. Uses the Bankr Submit API to execute transactions.

Create, edit, and refine agent skills through co-development and eval loops. Use for ANY question about skills or request to create/modify them.

The agent-native LLM router for OpenClaw. Route every request to the right model at the right price. 15-dimension scoring, 41+ models, zero API keys.

Monitors task execution for skill improvement opportunities. Use during any multi-step task, agentic workflow, or substantive work session where Claude is using tools and producing deliverables. Captures patterns, user corrections, workflow insights, and methodology worth preserving as reusable skills. Also triggers during post-task feedback, when the user mentions skill observations, improvements, the observation log, or asks Claude to watch for skill opportunities. Also known as "One Skill to Rule Them All."

name: test-driven-development description: Use when building a new feature, helper, API, or function and you want a test-first workflow. Follow a red-green-refactor loop: write one failing test for the smallest observable behavior, run to confirm it fails, implement only the minimal code to make it pass, then refactor code and test. ...

Sync tests with TestRail. Use when user mentions "testrail", "test management", "test cases", "test run", "sync test cases", "push results to testrail", or "import from testrail".

Automate NotebookLM notebook creation from YouTube videos. Given a YouTube video URL, extract information about people featured in the video, research them online, create a new NotebookLM notebook with the video and research as sources, and generate an audio overview. Uses screenshot-first automation to reliably target UI elements.

R3F performance optimization — LOD (Level of Detail), frustum culling, instancing strategies, draw call reduction, frame budgets, lazy loading, and profiling tools. Use when optimizing render performance, handling large or dense scenes, or debugging frame-rate issues.

Defines code quality benchmarks, standards compliance, and best practices for maintaining high-quality codebases

Box Least Squares (BLS) periodogram for detecting transiting exoplanets and eclipsing binaries. Use when searching for periodic box-shaped dips in light curves. Alternative to Transit Least Squares, available in astropy.timeseries. Based on Kovács et al. (2002).