发现 Skills

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution. Use when needing scalable evaluation on local Docker, Slurm HPC, or cloud platforms. NVIDIA's enterprise-grade platform with container-first architecture for reproducible benchmarking.

Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

This skill should be used when the user asks to "brainstorm research ideas", "use 5W1H framework", "identify research gaps", "conduct gap analysis", "start research project", "conduct literature review", "define research question", "select research method", "plan research", or mentions research project initiation phase. Provides comprehensive guidance for research startup workflow from idea generation to planning.

This skill should be used when the user asks to 'analyze skill quality', 'evaluate this skill', 'review skill quality', 'check my skill', or 'generate quality report'. Evaluates local skills across description quality, content organization, writing style, and structural integrity.

State-of-the-art text-to-image generation using Stable Diffusion models via the Hugging Face Diffusers library. Use this when generating images from text prompts, performing image-to-image translation, inpainting or outpainting, or building custom diffusion pipelines.

Statistical models library for Python. Use when you need specific model classes (OLS, GLM, mixed models, ARIMA) with detailed diagnostics, residuals, and inference. Best for econometrics, time series, rigorous inference with coefficient tables. For guided statistical test selection with APA reporting use statistical-analysis.

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Interactive planning for complex requests — QnA, plan file, structured approval.

デイリーノート朝用アシスタント（作成）

デイリーノートからKPT形式の月次振り返りを生成（引数なしで実行月、引数ありで指定月）

Use when you want Gemini CLI as a second opinion for coding tasks such as prototyping, debugging, or diff review, while keeping Codex as the primary implementer.

Automate Cdr Platform tasks via Rube MCP (Composio). Always search tools first for current schemas.

Automate New Relic tasks via Rube MCP (Composio): APM, alerts, dashboards, NRQL queries, and infrastructure monitoring. Always search tools first for current schemas.

Use this skill when starting or finishing substantial project work that would benefit from durable continuity across Codex, Claude, or other agent sessions. It trains agents to read recent project memory before work and to write a concise Markdown handoff under docs/memory/YYYY-MM-DD/project-or-task-memory-YYYY-MM-DD.md after meaningful work, including decisions, changed files, verifications, constraints, open questions, and next actions while avoiding secrets and sensitive details.

Generate focused project rules under .coddy/rules from repository analysis

Use after building an Android APK to visually verify and functionally test the app on a connected device or emulator via ADB. Installs the APK, navigates through screens, takes screenshots for visual analysis, sends touch/swipe inputs, checks logcat for crashes, and runs through a structured test script. Works with both physical devices and emulators.

Sub-agent powered code reviews spanning correctness, tests, consistency, and fit

Format markdown to GFM standard using oxfmt. Includes a structural guard (fence/table drift detection). Run after creating or editing any .md file. Uses the CLI at src/index.js or the bin entry "mdformat".

Part of the Blueprintflow methodology. Use on slow-cron ticks or when blueprint, current-doc, acceptance, PR, issue, or worktree drift signals appear.

Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations (for example: Codex, Responses API, Chat Completions, Apps SDK, Agents SDK, Realtime, model capabilities or limits); prioritize OpenAI docs MCP tools and restrict any fallback browsing to official OpenAI domains.

Read Reddit content via Composio MCP. Actions: search posts, view hot/top/new posts, read post content, read comments. Keywords: reddit, subreddit, post, comment, search reddit, hot posts, top posts.

Manages external API integrations for SunnahSleep. Use when working with Aladhan, ipwho.is, Nominatim, Open-Meteo, or Islamic Network CDN. Covers error handling, timeouts, fallbacks, and rate limiting.

—not trivia.