8,280 件 ・ 関連度順
概要と使いどころ
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
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。
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
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。
Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling. Use when training massive models across multiple machines or running distributed hyperparameter sweeps.
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。
Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.
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 for Vercel cost and performance optimization on deployed projects, especially Next.js, SvelteKit, Nuxt, and limited Astro apps. Collect Vercel metrics, usage, project config, and code scan results first; investigate only metric-backed candidates; produce ranked recommendations grounded in verified files and version-aware Vercel/framework docs. Trigger for Vercel bill reduction, slow or expensive routes, caching opportunities, Function Invocations, Build Minutes, Fast Data Transfer, Core Web Vitals, Bot Management, Fluid compute, or cost breakdown requests.
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 only changes since last commit using impact analysis. Token-efficient delta review with automatic blast-radius detection.
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。
为 FastGPT 新资源接入权限管理。当用户需要为新资源（如 AgentSkill、Plugin 等）添加权限支持时触发。
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。
Translate one, multiple, or all explicitly requested FastGPT i18next namespace JSON files from the completed Simplified Chinese source into every supported target locale, with product-language research and structural validation. Use only when the user explicitly invokes `$i18n-translate` and identifies namespace names, files under `packages/web/i18n/zh-CN/`, or all namespaces; never trigger implicitly for ordinary i18n, copywriting, or documentation work.
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。
为 FastGPT 新增、修改或审查自动系统升级脚本及其注册信息。涉及系统迁移、升级任务、启动迁移、migration registry、checkpoint、全量重跑、阻塞升级、进度或失败数据时使用；普通业务更新和未接入自动升级框架的一次性手工清洗不使用。
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。
手动触发的 FastGPT PR 或本地分支变更梳理技能。仅当用户显式调用 $pr-change-analysis 时使用；用于 reviewer 分析一个 GitHub PR 或当前本地分支相对 upstream/main 的需求变更、影响范围、代码质量与代码风格，不用于自动审查触发。
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。
FastGPT API 开发规范。重点强调使用 zod schema 定义入参和出参,在 API 文档中声明路由信息,编写对应的 OpenAPI 文档,以及在 API 路由中使用 schema.parse 进行验证。
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 prompt engineering skill that transforms Claude into "Alpha-Prompt" - a master prompt engineer who collaboratively crafts high-quality prompts through flexible dialogue. Activates when user asks to "optimize prompt", "improve system instruction", "enhance AI instruction", or mentions prompt engineering tasks.
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。
FastGPT CI workflow 双轨同步。当用户修改或新增 .github/workflows/ 下的 GitHub Actions workflow 时必须触发：同步更新 .forgejo/workflows/ 对应文件保持功能一致，或判断是否需要新建 Forgejo 版本。涉及 CI、GitHub Actions、Forgejo Actions、镜像构建、container registry、artifact、workflow yaml 改动、build-* workflow、test-* workflow 时也使用此技能。即使用户只提到"改一下 CI"或"加个 workflow"也应触发。
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。
将 FastGPT 文档从中文翻译为面向北美用户的英文。当用户提到翻译文档、i18n、国际化、translate docs、新增/修改了中文文档需要同步英文版时，使用此 skill。也适用于用户要求检查文档翻译缺失、批量翻译、或对比中英文文档差异的场景。
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。
当用户需要弃用一个工作流节点（保留向后兼容、隐藏出模板面板）时触发该 skill。FastGPT 工作流节点的弃用流程标准化封装，覆盖模板、Dispatcher、UI 引用等所有需要改动的位置。
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。
Runs the payroll-confidence chain end to end — forecasts cash across the payroll window with cash-flow-snapshot, ranks and drafts overdue-invoice reminders with invoice-chase to close any gap, then assembles timesheets, flags every anomaly, and stages the run with payroll-prep. On Gusto the run is staged for the owner to submit; on QuickBooks Payroll, or with no payroll system, the last step is a validated run sheet the owner keys in. This chain never presses the payroll button. Builds on the ledger (MYOB, NetSuite, QuickBooks, Xero, or Zoho Books), with Gmail or M365, PayPal, Square, Stripe, and either Gusto or QuickBooks Payroll adding detail, and falls back to CSV and timesheet uploads. Use it whenever payroll is the worry, including "can I make payroll," "payroll is due Friday," "plan payroll," "will I have enough to pay the crew," or "I need to run payroll but money is tight." Accepts optional horizon and payroll-date arguments.
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。
Makes sure no inbound lead goes unanswered. Watches web forms, the shared inbox, and the CRM for new inquiries, qualifies each one against the owner's criteria, drafts a reply in the owner's voice with real meeting times for the owner to approve and send, routes anything hot to a human right away, and logs the whole thing to the CRM. Never sends on its own. Built to run on a schedule so overnight inquiries are drafted and sorted by morning, and works from forwarded or pasted inquiries when no connector is available. Use this whenever the owner is worried about inbound leads or response time — including phrasings like "leads are going cold," "did anyone answer that inquiry," "I keep missing form submissions," "nothing falls through the cracks," "handle my inbound," "respond to this lead," or "we lose people because nobody gets back fast enough." Reach for it for a single unanswered inquiry as readily as for a standing setup.
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。
Finds grant and solicitation opportunities the organization actually qualifies for, runs a hard go/no-go before anyone starts writing, drafts the application or response from past submissions and real program data, and tracks every deadline and reporting obligation that follows an award. Built for nonprofits, government contractors, and education, and it filters daily feeds like SAM.gov or DIBBS down to the handful worth bidding. Runs on Google Drive or M365 with DocuSign and Calendar deepening it, and works from uploaded past submissions plus web research when nothing is connected. Use this whenever a funding or bidding opportunity is in play — including "find grants we qualify for," "should we bid this," "draft the response to this RFP," "we've got 50 solicitations to filter," "what's due on the federal grant," or "help me write the proposal narrative."
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。
Turns a plain-English description of a recurring report into a real, repeatable report — defines the metrics, pulls them from whatever data sources are connected, delivers a chat summary plus an XLSX workbook, and saves the definition so the same report reruns on demand or on a schedule. Handles anything from "sales by location versus last year" to AR aging to labor as a percentage of revenue. Works fully from an uploaded CSV or XLSX when no connector is available. Use this whenever the owner asks for a report, dashboard, KPI pack, metrics summary, or recurring numbers — including phrasings like "can you track this every week," "I need to see these numbers monthly," "build me a report on," "same report as last time," or "put together a KPI dashboard." Reach for it even when the owner names the metrics without using the word report.
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。
Reconciles the accounting ledger (MYOB, NetSuite, QuickBooks, Xero, or Zoho Books) against PayPal, Shopify, Square, and Stripe settlements, flags transactions that need attention, suspicious duplicates, and missing receipts, then writes a plain-English P&L narrative and exports a close packet (xlsx + one-page PDF). This is the first link of the /close-month command; a request to close the month or the books routes there, and the command runs this skill before refreshing the forecast and distributing the packet. Use this skill directly only when the owner wants the reconciliation alone, with no forecast refresh and no distribution: "just reconcile, no packet," "what's missing from the books," "flag the duplicates and missing receipts," or "write the P&L narrative for this month."
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。
Reads AR/AP, historical cash timing, and known fixed costs from the ledger (MYOB, NetSuite, QuickBooks, Xero, or Zoho Books) or from PayPal, Square, or Stripe — or a CSV upload — and produces a 30/60/90-day cash flow forecast with percentage-variance confidence bands and named risk flags. Delivers a chat summary and a downloadable XLSX. Use when the user asks "forecast my cash flow," "will I make payroll," mentions "runway," or says "cash crunch." Falls back to CSV upload when no connector is live.
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 nf-core bioinformatics pipelines (rnaseq, sarek, atacseq) on sequencing data. Use when analyzing RNA-seq, WGS/WES, or ATAC-seq data—either local FASTQs or public datasets from GEO/SRA. Triggers on nf-core, Nextflow, FASTQ analysis, variant calling, gene expression, differential expression, GEO reanalysis, GSE/GSM/SRR accessions, or samplesheet creation.
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。
Search agentmemory for past observations, sessions, and learnings about a topic using hybrid BM25 plus vector plus graph search. Use when the user says "recall", "what did we do about", "did we ever", "have we seen", or needs context from past sessions.
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。
Investigate a GitHub issue or reported bug the way a seasoned maintainer would — form an independent model before reading the thread's theories, trace the mechanism and history, verify the diagnosis, and return a concise briefing with a verdict and fix direction. Use for triage, root-cause analysis, or re-investigation.
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。