发现最适合你需求的 AI 技能
将 Agent 已解读好的正文写入 Qdrant（kb_main）。仅做 chunk、embedding 和向量写入，不负责抓取与精炼。
DeepSeek API 完整指南 - 注册、配置、省钱技巧。适合：想用便宜 AI 的用户。
获取干净、可读的网页正文内容，适合现代网页、博客、新闻、公告和微信公众号文章抓取；支持网页正文提取、内容清洗、去噪、Markdown 输出，适用于普通 fetch 效果不佳、页面噪音较多或动态渲染干扰的场景。Clean content fetch for modern web pages, article extraction, WeChat article capture, content cleanup, noise reduction, and markdown output when ordinary fetch is not clean enough.
Run and monitor PapersFlow DeepScan jobs. Use when the user wants long-running research progress, intermediate findings, final reports, or plotting from a completed run.
Preflight security scanner for AI coding agents — scans deployment config, skills/MCP servers, memory/sessions, and AI agent config files (hooks injection) for secrets, PII, prompt injection, and dangerous patterns. Runs 4 model behavior probes (persuasion, sandbagging, deception, hallucination). Supports LLM-enhanced semantic analysis. Works with OpenClaw, Claude Code, Cursor, and Codex. Use when a user asks for a security audit, health check, or wants to scan their AI agent setup for vulnerabi
End-to-end voice workflow with Deepgram STT and TTS. Use when transcribing voice messages, generating spoken replies, or building a shell-based audio pipeline that turns input audio into text and optionally returns an MP3 reply. Especially useful for Telegram/QQ/OneBot voice-message automation and Chinese speech workflows.
Qualify, prioritize, and advance sales prospects into the next best action. Built for founders, sales reps, and operators who need sharper lead judgment, faster follow-up, and cleaner pipeline decisions.
Two-layer memory architecture for board meeting decisions. Manages raw transcripts (Layer 1) and approved decisions (Layer 2). Use when logging decisions after a board meeting, reviewing past decisions with /cs:decisions, or checking overdue action items with /cs:review. Invoked automatically by the board-meeting skill after Phase 5 founder approval.
Create or refactor code for solving HJB equations with this repository's TensorFlow DGM framework. Use when users ask to generate new HJB training code, add a new problem (config/problem/loss/train script), adapt sampling/training hyperparameters, or create plotting/analysis code from trainer CSV outputs.
模型推荐 + OpenClaw 安全系数分析工具。当用户询问"用哪个模型"、"推荐模型"、"适合什么模型"、"安全系数"、"openclaw 安全"、"根据历史"时触发。功能：(1) 根据历史会话自动分析任务类型并推荐最优的 Claude 模型（auto 模式），(2) 根据用户描述的任务推荐最优模型（recommend 模式），(3) 读取本地 openclaw.json 计算当前网关的安全系数（0-100分）并给出评级。不适用于：模型切换操作、API Key 管理、套餐计费等。
|
Search, enrich, and manage family heritage and historical relationship data from the LiùDào (六道) SQLite database. Use this skill when answering questions about historical figures, calculating relationship paths (A的B), or adding new personal records with privacy controls.
