「aml」の検索結果

执行模型测评（统一入口）。LLM/VLM 走主流程 scripts/run_evaluation.py + Evaluation YAML；model_family=dit 走 DiT 扩展节（references/dit/evaluate_workflow.md，vbench.py 批量推理产出视频）。

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。

当用户直接要求进行数据集压缩测评，或在量化调优过程中需要使用 AISBench 评测工具输出数据集精度等场景时使用。 本 Skill 负责：使用 AISBench 数据集压缩代码、用 RBF Kernel Herding 算法生成压缩后的数据子集（coreset），并按用户目标交付两种产物之一： ① 接入量化调优主流程：产出独立的 coreset 数据集配置（不覆盖原始数据），回传「全集 config_name + coreset config_name」给编排层，供 evaluation.yaml 切换； ② 仅数据集压缩测评（不做量化调优）：基于 example 模板生成「全集测试脚本」和「子集测试脚本」两份 shell 脚本。 暂时支持 AIME 2025 和 GPQA，其他数据集需要再结合 aisbench tools目录下 herding_coreset_selector 源码进行分析。

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。

Append a DiT-tuning round to history.yaml. DiT-side counterpart of the LLM-path accuracy_append.py; records practice.md5, inference_outputs, fp_baseline_outputs, and the scoring fields (scores / overall_score / loss_vs_baseline / is_satisfied) populated by quant-tuning-score-dit. Idempotent on practice_id.

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。

On-brand visual marketing content using Paper MCP and parallel agent teams. Reads the project's brand/ directory to build a design system, then spawns designer agents in parallel — each creating a unique artboard with a different on-brand layout interpretation. Intelligently adapts agent count and approach based on user goals via AskUserQuestion interrogation. Produces Instagram carousels, TikTok slideshows, social posts, story slides, and visual assets. When used with content spec YAMLs from /slideshow-script, each agent gets a unique script AND unique design direction. Make sure to use this skill whenever the user wants visual marketing content designed in Paper — carousels, slideshows, social posts, story slides, or any visual asset. Even if they just say "design something" or "make slides" or "create visual content", this is the skill. Also triggers on "paper marketing", "instagram design", "TikTok design", "slideshow design", "create carousel", or "social graphics".

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。

Generate 5 different narrative scripts for visual slideshows from a single positioning angle. Each script uses a different storytelling framework (AIDA, PAS, BAB, Star-Story-Solution, Stat-Flip) producing genuinely different stories, not layout variations. Make sure to use this skill whenever the user wants slideshow scripts, TikTok content scripts, carousel copy, narrative frameworks for visual content, or says anything about writing scripts for slides or social media storytelling. Even 'write me some TikTok content' or 'I need carousel copy' should trigger this. Outputs structured YAML content specs that chain directly to /paper-marketing for visual design.

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。

Set up or refine open-source .NET code-quality gates for CI: formatting, `.editorconfig`, SDK analyzers, third-party analyzers, coverage, mutation testing, architecture tests, and security scanning. Use when a .NET repo needs an explicit quality stack in `AGENTS.md`, docs, or pipeline YAML.

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。

Datadog provider for obz. Supports metrics (Datadog Query Language), logs (Datadog Log Query), and traces. Requires api-key and app-key configured under auth in config.yaml. Does not support metric labels, label-values, or series commands. This skill should be used when the user mentions "Datadog", "DD", "obz -p dd", Datadog Query Language, or needs to configure Datadog provider instances with regional endpoints.

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。

Deploy applications to Render by analyzing codebases, generating render.yaml Blueprints, and providing Dashboard deeplinks. Use when the user wants to deploy, host, publish, or set up their application on Render's cloud 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。

Generates Claude Code plugin agent .md files: writes YAML frontmatter (name, model, color, tools, hooks, disallowedTools), crafts system prompts, and creates triggering example blocks with context and commentary. Validates agent structure against naming, length, and format constraints. Use when the user asks to create an agent, add an agent, write a subagent, configure agent frontmatter, design agent system prompts, set agent tools or colors, build an autonomous agent, or add agent hooks.

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。

Claude Code agent generation system that creates custom agents and sub-agents with enhanced YAML frontmatter, tool access patterns, and MCP integration support following proven production patterns

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。

Generate AGENT.md files with proper YAML frontmatter, role definitions, expertise areas, and prompt templates following Babysitter SDK conventions.

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 "create an agent", "write a subagent", "generate agent definition", "add agent to plugin", "write agent frontmatter", "create autonomous agent", "build subagent", needs agent structure guidance, YAML frontmatter configuration, invocation criteria with examples, or wants to add specialized subagents to Claude Code plugins with proper capabilities lists and tool access definitions.

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。

Create, update, review, fix, or debug VS Code agent customization files (.instructions.md, .prompt.md, .agent.md, SKILL.md, copilot-instructions.md, AGENTS.md). Use for: saving coding preferences; troubleshooting why instructions/skills/agents are ignored or not invoked; configuring applyTo patterns; defining tool restrictions; creating custom agent modes or specialized workflows; packaging domain knowledge; fixing YAML frontmatter syntax.

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。

Creates Claude Code agent files (.claude/agents/*.md) with YAML frontmatter, tool permissions, and system prompts. Use when building new agents, configuring agent frontmatter, writing agent system prompts, or setting agent tool permissions.

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。

Cleans a Quarto (.qmd) file by replacing LaTeX holdovers with proper Quarto/Pandoc syntax so the document renders correctly to HTML, PDF (Typst/LaTeX), and Word. Fixes citations (\citep → [@key]), cross-references (\ref → @label), figures (\includegraphics → markdown), tables, text formatting (\textbf → **bold**), inline math (\( \) → $ $), hyperlinks (\href → [text](url)), footnotes (\footnote → ^[]), list environments, section headings, and R Markdown chunk options (dot-style → #| YAML). Use when asked to clean, fix, modernize, or port a Quarto or R Markdown document; when LaTeX commands appear in a .qmd file; when a document fails to render to Word or Typst; or when citations/cross-references do not appear in non-LaTeX output.

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。

Converts LaTeX article-class .tex documents to Quarto .qmd format for multi-format publishing (HTML, PDF, Word). Use when asked to convert, port, migrate, or translate a .tex LaTeX file to Quarto; when porting an academic paper or manuscript from LaTeX to Quarto; when a .tex document needs to render to HTML, Word, or Typst PDF. Covers preamble → YAML front matter, section headings, text formatting, math, citations (natbib/biblatex), cross-references, figures, tables, lists, footnotes, hyperlinks, and special characters.

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。

"Apply the SERVQUAL model (Parasuraman, Zeithaml, and Berry, 1988) to measure service quality gaps across five dimensions. Use this skill when the user needs to diagnose service quality shortfalls, benchmark customer expectations against perceptions, design service improvement programs, or when they ask 'where is our service failing', 'what do customers expect vs experience', or 'how do we measure service quality'.".

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。

"Apply Lean and Six Sigma principles to eliminate waste and reduce process variation. Use this skill when the user needs to improve operational efficiency, identify the seven wastes (TIMWOOD), run DMAIC improvement projects, or streamline workflows — even if they say 'our process is too slow', 'where are we wasting resources', or 'how do we reduce defects'.".

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。

Searches for Memorie van Toelichting (explanatory memoranda) for a Dutch law and generates Gherkin test scenarios from legislature-intended examples. Use this skill proactively when: user wants MvT-derived BDD scenarios, mentions "memorie van toelichting", "MvT", "parlementaire stukken", "rekenvoorbeelden", or "kamerstukken" in the context of Dutch law. Activate automatically when a new law YAML file is created and the user discusses testing or validation.

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。

Compare policies, procedures, onboarding files, or operating evidence against a named compliance framework such as AML/KYC, privacy, internal controls, or audit readiness. Use when a team needs a gap matrix, control checklist, remediation priorities, or evidence request list. Do not use when no target framework is specified or when formal legal or regulatory sign-off is required.

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。

Execute advanced Man-in-the-Middle (MITM) NTLM Relay attacks specifically targeting environments where SMB Signing is purportedly enabled but improperly configured (e.g., exclusively required on Domain Controllers but NOT enforced universally on standard Workstations or File Servers). Utilize powerful tools like Responder and Impacket's NTLMRelayX to coercively capture NTLMv2 authentications (via LLMNR, WPAD, or Coercion methods like PetitPotam) and seamlessly relay them across the network identically yielding immediate Local Administrator command execution.

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。

Abuse Azure Managed Identities from compromised Azure Virtual Machines (VMs), Functions, or App Services to seamlessly request valid, highly-privileged Azure AD access tokens and laterally move throughout the cloud environment without requiring explicit credentials.

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。

Exploit Active Directory Certificate Services (AD CS) misconfigurations, specifically ESC1. By requesting a certificate based on a overly permissive template that allows the enrollee to supply a Subject Alternative Name (SAN), an attacker can impersonate highly privileged users (like Domain Admins) and seamlessly escalate privileges across the entire AD environment.

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。

Review computational protein or gene function predictions and fill PredictionReview YAML files using the COR/CNN/LSP/UNC/PLI/NPI/REP biological-validity taxonomy. Use for ProtNLM, DeepECTF, BioReason/GO-GPT, InterPro2GO, PANTHER/IBA, CLEAN, GloEC, MAPred, ProteinInfer, or other predicted EC/GO annotations, including audits of the evidence and reasoning in existing prediction reviews.

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。