データ・AIのスキル

Build production-ready LLM applications, advanced RAG systems, and intelligent agents. Implements vector search, multimodal AI, agent orchestration, and enterprise AI integrations.

日本語の概要は準備中です。原文の説明を表示しています。

This skill should be used when the user asks to "implement LLM-as-judge", "compare model outputs", "create evaluation rubrics", "mitigate evaluation bias", or mentions direct scoring, pairwise comparison, position bias, evaluation pipelines, or automated quality assessment.

日本語の概要は準備中です。原文の説明を表示しています。

permission-ledger hook が記録した permission prompt と各リポの .claude/settings.local.json から allow 昇格候補を集計し、ユーザーが選んだものだけを ai-agents/settings/claude/settings.json に反映する。 「許可ルールを見直したい」「permission prompt を減らしたい」「承認した操作を allow に入れたい」 「permission review」などと言われたときに `/permission-review` で使う。

Android data layer implementation guide. Provides comprehensive patterns for Room Database (Entity, DAO, Migration), DataStore (Preferences), Repository layer design, and common pitfalls. Use when implementing local data storage, cache strategies, or data source integration in Android projects.

日本語の概要は準備中です。原文の説明を表示しています。

Refresh the Key Metrics table in a thesis with current data from Financial Modeling Prep, with a scoped web-search fallback (Step 4b) when FMP has no data or resolved the wrong company. Use when user says "numbers", "refresh metrics", "update key metrics", or "refresh [TICKER]" with no other action implied. Surgical edit only — does NOT create research notes or propagate via $sync. Also flags (never auto-edits) stale price/valuation framing in the thesis Summary — see Step 10b.

日本語の概要は準備中です。原文の説明を表示しています。

Refresh the Key Metrics table in a thesis with current data from Financial Modeling Prep, with a Quartr standardized-statement tie-break for Revenue Growth / Gross Margin gaps (Step 4b-Q) and a scoped web-search fallback (Step 4b) when FMP has no data or resolved the wrong company. Use when user says "numbers", "refresh metrics", "update key metrics", or "refresh [TICKER]" with no other action implied. Surgical edit only — does NOT create research notes or propagate via /sync. Also flags (never auto-edits) stale price/valuation framing in the thesis Summary — see Step 10b.

日本語の概要は準備中です。原文の説明を表示しています。

Download, screen, and map non-North-America auxiliary EcoSIM climate variables from global or regional atmospheric-chemistry data sources. Use when a non-US/non-NLDAS site needs annual EcoSIM precipitation chemistry variables such as PHRG, CN4RIG, CNORIG, CSORG, CCARG, CMGRG, CNARG, CKARG, or CCLRG; when deciding whether to use CAMS, EBAS, EANET, or MERRA-2; or when deriving site-specific auxiliary variables after ERA5 climate forcing has been prepared.

日本語の概要は準備中です。原文の説明を表示しています。

Prepare, validate, inspect, and edit EcoSIM soil management inputs using soil_mgmt_in NetCDF, editable Excel workbooks, and optional JSON files. Use when checking fertilizer, tillage, or irrigation management data; converting soil-mngt NetCDF to Excel for editing; converting edited Excel back to JSON or NetCDF; decoding fertf/tillf/irrigf selector variables; or validating EcoSIM ReadManagementMod.F90 soil-management inputs.

日本語の概要は準備中です。原文の説明を表示しています。

Extract dominant-component soil profile variables for EcoSIM grid or template inputs from gSSURGO, with FAO HWSD v2.0 fallback when gSSURGO data are unavailable or incomplete. Use when deriving soil depth, bulk density, field capacity, wilting point, hydraulic conductivity, texture, rock fraction, pH, CEC, or soil organic carbon; or when converting EcoSIM grid NetCDF files to editable Excel workbooks and back.

日本語の概要は準備中です。原文の説明を表示しています。

Download NASA GES DISC NLDAS_FORA0125_H version 2.0 primary hourly forcing data for a point location. Use when Codex needs to retrieve NLDAS File A meteorological forcing from the GES DISC API or HTTPS archive for a longitude/latitude pair, including Earthdata username/password authentication, hourly URL generation, nearest-grid-point extraction, CSV output, or EcoSIM climate-forcing preprocessing.

日本語の概要は準備中です。原文の説明を表示しています。

Run the combined AmeriFlux to EcoSIM extraction workflow for site metadata, NADP or tDEP atmospheric chemistry, gSSURGO soil data, and optional climate or grid NetCDF forcing. Use when building a complete site-specific EcoSIM input package.

日本語の概要は準備中です。原文の説明を表示しています。

Prepare, validate, and edit EcoSIM plant management inputs using pft_mgmt_in NetCDF, editable Excel workbooks, and optional PlantMgmtWriter.py JSON files. Use when creating or checking natural PFT management data, converting pft-mngt NetCDF to an Excel sheet for editing, converting edited Excel back to JSON or NetCDF, validating pft_type/pft_pltinfo/pft_mgmt entries, or preparing PlantMgmtWriter.py inputs for EcoSIM.

日本語の概要は準備中です。原文の説明を表示しています。

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

日本語の概要は準備中です。原文の説明を表示しています。

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image.

日本語の概要は準備中です。原文の説明を表示しています。

Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework

日本語の概要は準備中です。原文の説明を表示しています。

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

日本語の概要は準備中です。原文の説明を表示しています。

Activate when the user asks about any repository listed in the system prompt under 'OpenViking — Indexed Code Repositories', or when they ask about an external library, framework, or project that may have been indexed. Also activate when the user wants to add, remove, or manage repos. Always search the local codebase first before using this skill.

日本語の概要は準備中です。原文の説明を表示しています。

Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods, 28 analysis modules, 116 model presets across 5 compute tiers, tournament evaluation, and telemetry-driven recommendations. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.

日本語の概要は準備中です。原文の説明を表示しています。

This skill should be used when the user wants to visualize data. It intelligently selects the most suitable chart type from 26 available options, extracts parameters based on detailed specifications, and generates a chart image using a JavaScript script.

日本語の概要は準備中です。原文の説明を表示しています。

Jailbreak API-served LLMs using G0DM0D3 techniques — Parseltongue input obfuscation (33 techniques), GODMODE CLASSIC system prompt templates, ULTRAPLINIAN multi-model racing, encoding escalation, and Hermes-native prefill/system prompt integration. Use when a user wants to bypass safety filters, jailbreak a model, or red-team model robustness.

日本語の概要は準備中です。原文の説明を表示しています。

Use this skill when the user requests to generate, create, or write professional research reports including but not limited to market analysis, consumer insights, brand analysis, financial analysis, industry research, competitive intelligence, investment due diligence, or any consulting-grade analytical report. This skill operates in two phases — (1) generating a structured analysis framework with chapter skeleton, data query requirements, and analysis logic, and (2) after data collection by other skills, producing the final consulting-grade report with structured narratives, embedded charts, and strategic insights.

日本語の概要は準備中です。原文の説明を表示しています。

Use this skill when the user uploads Excel (.xlsx/.xls) or CSV files and wants to perform data analysis, generate statistics, create summaries, pivot tables, SQL queries, or any form of structured data exploration. Supports multi-sheet Excel workbooks, aggregation, filtering, joins, and exporting results to CSV/JSON/Markdown.

日本語の概要は準備中です。原文の説明を表示しています。

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.

日本語の概要は準備中です。原文の説明を表示しています。

Songwriting craft, AI music generation prompts (Suno focus), parody/adaptation techniques, phonetic tricks, and lessons learned. These are tools and ideas, not rules. Break any of them when the art calls for it.

日本語の概要は準備中です。原文の説明を表示しています。