AI & ML

dreamers-laboratory Bundle

Find every place agent skills live on this machine (Claude Code, Codex, Cursor; user-level and project-level) and bring them into sync, bidirectionally, newest copy wins. Use when the user asks to sync skills between tools, says a skill is missing or stale in one tool, wants to know which tools share a skills directory, or after editing a skill that other tools should see. Reports before it changes anything; backs up anything it overwrites; never deletes.

dreamers-laboratory Skill

Perform a general Linux security breach check and compromise audit. Use when the agent is asked to audit a Linux server for unauthorized access, bad actor SSH keys, suspicious users, privilege escalation, persistence, suspicious processes/services, network listeners, containers, writable temp-path activity, or post-compromise access cleanup via SSH, console, GRUB, initramfs, or rescue/live boot.

dreamers-laboratory Skill

Collaborate with OpenAI Codex, through its own app or CLI, as a respectful second-opinion partner for engineering, refactor review, analysis, verification, and gap analysis. Use when the user asks another agent to talk to Codex, hand work to Codex, learn from Codex's output, compare Codex's critique, get Codex's implementation or debugging ideas, or use Codex as a peer reviewer before finalizing an artifact.

dreamers-laboratory Skill

Collaborate with Claude, through its own app or CLI, as a respectful second-opinion partner for formatting, diagrams, document and deck review, writing, visual critique, and gap analysis. Use when the user asks another agent to talk to Claude, hand work to Claude, learn from Claude's output, compare Claude's critique, get Claude's formatting or design ideas, or use Claude as a peer reviewer before finalizing an artifact.

dreamers-laboratory Bundle

Read and analyze live OpenAI Codex account usage limits from the local authenticated app server. For users of the OpenAI Codex app/CLI only. Use when the agent needs to report current 5-hour or weekly usage, reset times, average weekly percentage used per elapsed day, sustainable percentage remaining per day, pace against the window, or available reset credits without opening the GUI.

dreamers-laboratory Skill

Create a secret-safe project context handoff file in /tmp for another model, agent, thread, or teammate. Use when the agent needs to summarize everything relevant to an overall objective across repositories, docs, dashboards, deployment systems, access paths, credentials, live services, open decisions, blockers, and next actions, especially before handing off complex work or preserving project state.

dreamers-laboratory Bundle

Source-grounded AI model comparison workflow for comparing frontier LLMs, coding agents, chatbots, and unreleased or preview models. Use when the agent needs to research model benchmarks, verify published model claims, build coding, general-task, or design benchmark tables, compare OpenAI/Anthropic/Google/xAI/Meta/Mistral/DeepSeek or similar models, update model-comparison content pages, or explain uncertainty around unpublished, gated, preview, or rumored models.

loulanyue Skill

Unified media generation via fal.ai MCP — image, video, and audio. Covers text-to-image (Nano Banana), text/image-to-video (Seedance, Kling, Veo 3), text-to-speech (CSM-1B), and video-to-audio (ThinkSound). Use when the user wants to generate images, videos, or audio with AI.

loulanyue Skill

Research-before-coding workflow. Search for existing tools, libraries, and patterns before writing custom code. Invokes the researcher agent.

loulanyue Skill

Interactive agent picker for composing and dispatching parallel teams

j2h4u Bundle

This skill should be used when the user asks to "design an MCP server", "audit an MCP server", "review MCP tools", "add MCP tool", "improve tool descriptions", "design tool surface", "add submit_feedback tool", "review tool schema", mentions "MCP transport", "tool annotations", "mcp stdio", "MCP Resources", "MCP Prompts", or is designing, reviewing, or auditing any MCP server. Covers design philosophy, tool naming, parameter schemas, agent UX, feedback channel, transport, security, and client compatibility. NOT for hands-on implementation from scratch (use mcp-builder for that).

lunarxuan Bundle

Reverse-engineer a supplied video into a shot-by-shot generation specification, displayed evidence frames, and paired Chinese-English prompt packages. Use for requests such as 视频反推提示词, 视频转提示词, 从视频生成提示词, 视频逆向提示词, 逐镜头反推, 分镜拆解并生成提示词, 反推视频 prompt, or video-to-prompt; also use for Veo, Sora, Kling, Wan, generic prompt adaptation, and bounded generate-compare-optimize loops. Do not use for ordinary video summaries or basic transcription.

omry Skill

API Demo Agent Skill

omry Bundle

Use this wheel-packaged demo skill to verify local wheel installs.

omry Bundle

Demo Agent Skill

warlcang Bundle

Run a mock technical interview about the current repository, in any language the user prefers. Use when the user asks to be interviewed, grilled, quizzed, or drilled about their project/repo/codebase (in English, 中文, or any other language), wants to prepare for a technical interview about their own project, needs to practice "walk me through your project" for an engineering, PM, design, or data role, wants to defend a portfolio or vibe-coded project they built with AI, or asks "do I actually understand this code". Reads the repo and (when available) local agent session logs, asks staff-engineer-quality questions one at a time, scores answers honestly with grounded coaching, and tracks a readiness score across sessions.

dreamers-laboratory Bundle

Compute required descriptive and inferential statistics for measured samples. Use when the agent is analyzing latency, throughput, event timing, experiment baselines, A/B effects, benchmark results, production-vs-baseline comparisons, or any numeric measurement where outputs must include sample count, min, max, mean, median, standard deviation, standard error when relevant, confidence intervals when useful, and t-tests for measured effects.

dreamers-laboratory Bundle

Find and digest recent research papers through the alphaXiv MCP server (discover_papers, get_paper_content, answer_pdf_queries, researcher and library tools), with an arXiv/OpenAlex fallback when the MCP is not configured. Token-aware by design. The agent frames a token budget per goal before reading anything, keeps a ledger, and reads abstracts before reports before full text. Offers OpenResearch (openresearch.sh) autoresearch as an option when the question is empirical and the user has code to run it against. Use when the user asks what is new on a topic, wants a literature scan, needs the latest method for a problem, wants a specific paper read or queried, or wants to find who works on something.

dreamers-laboratory Skill

Design, run, and evaluate controlled empirical experiments with one variable changed at a time, explicit hypotheses, reproducible protocols, statistical significance checks such as t-tests where appropriate, and conclusions backed by measured evidence. Use when the agent is planning or judging performance tests, latency experiments, A/B comparisons, benchmark runs, optimization claims, production-vs-baseline comparisons, or any investigation where assumptions must be minimized and results must be statistically defensible.

dreamers-laboratory Bundle

Design, implement, and empirically verify high-value automated regression tests against historical application defects. Use when the agent must mine issue history, cluster recurring failure modes, build code or data-integrity tests, and prove each claimed catch by failing on the historical defect while passing on corrected application code.

codesoul-co Bundle

Collects bounded public evidence and preserves source provenance.

loulanyue Skill

Head-to-head comparison of coding agents (Claude Code, Aider, Codex, etc.) on custom tasks with pass rate, cost, time, and consistency metrics

loulanyue Skill

Anthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.

loulanyue Skill

Neural search via Exa MCP for web, code, and company research. Use when the user needs web search, code examples, company intel, people lookup, or AI-powered deep research with Exa's neural search engine.

loulanyue Skill

Head-to-head comparison of coding agents (Claude Code, Aider, Codex, etc.) on custom tasks with pass rate, cost, time, and consistency metrics

loulanyue Skill

Anthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.

dreamers-laboratory Bundle

Fill, stamp, or prepare PDF forms only when visual alignment matters. Use when the agent needs to add text to PDF forms, prefill government/vendor PDFs, prepare sign-ready packets, repair bad PDF fills, center text in boxes, or verify that generated PDFs look correct from rendered page screenshots before sharing or attaching them.

mattlgroff Bundle

Audits MCP servers that an organization builds or operates, including implementations and launch roadmaps, against the current MCP specification and provider-side OWASP controls. Use when reviewing a served MCP endpoint, its tools, resources, prompts, transport, authorization, DCR or CIMD registration, deployment, logging, readiness, or launch plan. Do not use for selecting or governing third-party MCP servers an organization consumes.

loulanyue Skill

Neural search via Exa MCP for web, code, and company research. Use when the user needs web search, code examples, company intel, people lookup, or AI-powered deep research with Exa's neural search engine.

loulanyue Skill

Head-to-head comparison of coding agents (Claude Code, Aider, Codex, etc.) on custom tasks with pass rate, cost, time, and consistency metrics

loulanyue Skill

Anthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.

loulanyue Skill

Neural search via Exa MCP for web, code, and company research. Use when the user needs web search, code examples, company intel, people lookup, or AI-powered deep research with Exa's neural search engine.

loulanyue Skill

Scan your Claude Code configuration (.claude/ directory) for security vulnerabilities, misconfigurations, and injection risks using AgentShield. Checks CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions.

loulanyue Skill

编码代理（Claude Code、Aider、Codex等）在自定义任务上的直接比较，包含通过率、成本、时间和一致性指标

loulanyue Skill

Anthropic Claude API 的 Python 和 TypeScript 使用模式。涵盖 Messages API、流式处理、工具使用、视觉功能、扩展思维、批量处理、提示缓存和 Claude Agent SDK。适用于使用 Claude API 或 Anthropic SDK 构建应用程序的场景。

loulanyue Skill

通过Exa MCP进行神经搜索，适用于网络、代码和公司研究。当用户需要网络搜索、代码示例、公司情报、人员查找，或使用Exa神经搜索引擎进行AI驱动的深度研究时使用。