2,257 件 ・ 関連度順
概要と使いどころ
Plataforma de observabilidade para LLM com rastreamento, avaliação e monitoramento. Use ao depurar aplicações LLM, avaliar saídas de modelos em datasets, monitorar sistemas em produção ou construir pipelines de testes sistemáticos para aplicações de IA.
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。
Fornece orientação para realizar intervenções causais em modelos PyTorch usando o framework de intervenção declarativa do pyvene. Use ao conduzir rastreamento causal, activation patching, treinamento de intervenção de intercâmbio ou testar hipóteses causais sobre o comportamento do modelo.
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。
Crie e gerencie arquivos de planejamento markdown persistentes para execução estruturada de tarefas. Use quando o usuário pedir para "criar um plano", "acompanhar progresso", "iniciar um projeto de pesquisa", ou quando uma tarefa exigir mais de 5 chamadas de ferramenta e precisar de rastreamento de fases estruturado para manter o foco e evitar desvio de objetivo.
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。
Processamento escalável de dados para workloads de ML. Execução em streaming em CPU/GPU, suporta Parquet/CSV/JSON/imagens. Integra-se com Ray Train, PyTorch, TensorFlow. Escala de uma única máquina para centenas de nós. Use para inferência em lote, pré-processamento de dados, carregamento de dados multi-modal ou pipelines distribuídos de ETL.
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。
Construa pipelines de dados escaláveis, data warehouses modernos e arquiteturas de streaming em tempo real. Implementa Apache Spark, dbt, Airflow e plataformas de dados nativas em nuvem.
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。
Avalia LLMs em mais de 60 benchmarks acadêmicos (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use quando benchmarking de qualidade de modelo, comparação de modelos, relatório de resultados acadêmicos ou rastreamento de progresso de treinamento. Padrão da indústria usado por EleutherAI, HuggingFace e laboratórios principais. Suporta HuggingFace, vLLM, APIs.
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。
[Code Quality] Use to validate a spec artifact's decisions with the user across three contexts — a freshly-authored Feature Spec (idea-to-spec), an existing canonical spec before PBI decomposition (spec-to-pbi), or a refined idea + §8 test-specs (idea-to-pbi deep mode). Detects the context, walks every applicable validation category, and runs an exhaustive but budget-bounded blocking clarification gate so every non-obvious or conflicting decision is confirmed before the artifact drives downstream 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。
Install, discover, and wire registry blocks and components into HyperFrames compositions. Use when running hyperframes add or hyperframes catalog, installing one item or every block matching a tag, wiring an installed item into index.html, or working with hyperframes.json. Covers discovery, install locations, block sub-composition wiring, component snippet merging, and authoring a new block or component to contribute upstream (idea → scaffold → validate → PR).
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。
Collect project brief and stack choices, then output canonical project-config.json for downstream web-builder skills.
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。
Define legal scope, rules of engagement, and authorization artifacts for downstream pentest 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。
Design, audit, harden, review, or upgrade MCP servers and MCP Apps with revision-aware protocol profiles, explicit HTTP/deployment boundaries, capability-centered validation, bilingual reporting, and reproducible evidence. Use for MCP server architecture, JSON-RPC or Streamable HTTP bugs, security hardening, protocol-version migration, MCP Apps widget boundaries, review reconciliation, test planning, and public engineering evidence.
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。
Pull the latest GEO-SEO skill updates from the upstream repository. Compares installed files against the latest release, shows what changed, and updates all skills, agents, scripts, and schema templates in place.
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。
GStack retrospective method for completed workstreams, identifying what worked, what failed, process changes, and role-skill improvements.
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。
GStack spec method for turning fuzzy intent into an executable spec with goals, non-goals, acceptance criteria, risks, and downstream role boundaries.
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。
GStack plan-tune method for calibrating how many questions to ask, autonomy level, stop conditions, and decision style for a workstream or user preference.
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。
GStack guard method for strengthening scope boundaries, forbidden actions, stop conditions, and validation gates for an agent or workstream.
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。
Validate and implement Sinclair ZX Spectrum 48K ROM-visible behavior in the EMULATORS repository, including reset and restart vectors, IM1/NMI handling, keyboard scanning and repeat, beeper and tape routines, screen/printer output, channels and streams, BASIC tokenization and execution, expression evaluation, memory movement, system variables, calculator stack, floating-point formats, ROM traps, localized ROM selection, and firmware/hardware timing contracts. Use when Codex reviews or edits ZX Spectrum ROM loading, KERNAL-style traps, keyboard, tape, screen, printer, BASIC files, tokens, system variables, startup, interrupts, or compatibility tests.
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。
SOP for the architect-tier campaign workflow (/campaign): exhaustive survey, mitigation planning, tiered orchestration, and reconciliation. Trigger when: - Running a multi-workstream initiative where an expensive architect-tier council surveys, plans, emits worker prompts, and judges landed work. - Conducting production-readiness assessments that fan out into autonomous mitigation dispatches across model tiers. - Prompt contains: /campaign, campaign workflow, survey, orchestrate, reconcile, premise freshness, tier routing, worker IBC, scratch.
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。
Discipline for repairing a false claim once it has propagated: select what depends on it, produce a replacement, and accept with an expiring verdict. Trigger when: - A claim, citation, premise, or documented fact is found wrong and its downstream consequences must be traced and fixed. - A refutation, a failed re-run, a retracted external source, or a human correction lands against something already closed or already promoted. - Prompt contains: /remediation, remediation, retraction, false claim, propagated error, scrub, correction propagation, born-false, became-false, currency verdict, impeachment queue.
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。
Efficiently extract and convert the contents of any local file—such as PDF, DOCX, DOC, ODT, RTF, XLSX, XLS, or HTML—into clean, well-formatted markdown saved to disk. Use this skill whenever the user requests to parse, read, or extract information from a file on their computer, including phrases like “parse this PDF”, “convert this document”, “read this file”, “extract text from”, or when a local file path (not a URL) is provided. This skill offers advanced options like generating AI-powered summaries and answering questions based on the file's content. Prefer this tool over `scrape` when handling local files to deliver precise, structured outputs for downstream 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。
Use for DESIGN.md visual identity contracts; author, scaffold, lint, diff, and export design contracts with the upstream spec and CLI.
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 notebook-connector configuration for Exasol AI workflows through the Secrets Python API. Covers secure config store (`scs`) creation, the `AILabConfig` and `StorageBackend` key sources, backend-specific config values such as `db_host_name`, `db_schema`, `storage_backend`, and `huggingface_token`, Python validation, and handoff to downstream notebook-connector skills.
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。
Efficiently bring upstream NanoClaw updates into a customized install, with preview, selective cherry-pick, and low token usage.
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。
Check for and apply updates to installed skill branches from upstream.
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。