1.8万 件 ・ 関連度順
概要と使いどころ
GEODE evidence-first development workflow. Use for feature work, provider/model changes, GUI/computer-use work, PDF/document ingestion, observability/schema/event/state changes, verification planning, GitFlow merge prep, and requests to "follow the workflow" or "scaffold the process".
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。
Implement proven backend architecture patterns including Clean Architecture, Hexagonal Architecture, and Domain-Driven Design. Use when architecting complex backend systems or refactoring existing applications for better maintainability.
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。
Align, run, audit, and publish GEODE benchmark comparisons against Agent-World arXiv 2604.18292v1 across MCP-Mark, BFCL V4, and tau2-bench. Use for Agent-World or AgentWorld comparison requests, cross-suite agent benchmark design, Tau2 comparator/profile diagnosis, paired thin-runtime-versus-GEODE experiments, mean-accuracy@8 replication, or benchmark comparability and artifact 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。
Audit and improve Generative Engine Optimization as a measurable discovery-to-citation pipeline. Use for /geo, AI-search visibility, citation readiness, and GEO benchmark 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。
Review all unreviewed documents in Hubdoc. Use when reviewing, processing or publishing Hubdoc receipts and bills.
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 a pull request description for the current branch and writes it as Markdown to .git/magit/posts/new-pullreq for magit-forge. Use when the user asks for a PR description, PR summary, or write-up of what a branch changes, or is preparing to open a pull request.
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。
Applies local reliability, viewport, tab-safety, stale-ref, widget, and browser-boundary lessons to every agent-browser automation. Use alongside the upstream agent-browser skill whenever navigating, clicking, filling, testing, extracting, taking screenshots, or debugging any website with agent-browser.
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。
Refactor code units whose size or mixed responsibilities impede understanding, testing, or change. Use when recent work makes a function, class, or module materially harder to reason about.
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。
Upload images and files to GitHub PRs, issues, and comments so GitHub hosts them instead of committing binaries to the repository. Use when adding screenshots to a PR or issue body, replacing committed screenshots or raw.githubusercontent.com links with hosted attachments, or when `gh` appears to offer no way to attach a file.
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 changed code for correctness, security, maintainability, and verification gaps. Use when code is ready for independent review or before handoff, staging, or merge.
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。
Triages an already-dirty Git working tree: breaks down every uncommitted change hunk by hunk, separates deliberate edits from tool-written churn, possibly accidental reversions, misplaced scope, and sensitive content, asks the user about the ambiguous cases, and proposes an ordered set of commits plus a list of hunks to hold back. Use when the user invokes `/pc` or `$pc`, asks what they should commit, wants their uncommitted changes broken down, or wants commits proposed for a dirty working tree or pile of accumulated local modifications. To plan commit boundaries for a change before or while implementing it, use `incremental-commits` instead.
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。
Rebases every local branch of a fork onto a moved upstream, flags the ones upstream has superseded, and rebuilds the mixdown branch that combines them. Use after fetching a fork's upstream and finding it has advanced far, been force-pushed, or rewritten its history; when several stacked or sibling branches all need bringing forward at once; when a branch was cut from a mixdown and now carries merge commits it should not have; or when asked to rebase all branches, re-mix, or bring a whole worktree set up to date.
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。
Used by the Reviewer agent during Phase 9 (DesignVerify) and Phase 11 (Quality) to evaluate frontend code. Assesses visual hierarchy, spacing consistency, accessibility, and detects formulaic 'AI slop' design patterns, providing actionable feedback to the Developer.
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 database migrations safely in production. Zero-downtime strategies, rollback procedures, backward compatibility.
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 TUI screen/view with layout, keybindings, state management, and 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。
Revisa o produce textos visibles para la persona usuaria (texto de UI, encabezados, botones, mensajes de error, marketing) con la voz y el tono del proyecto. Úsalo cuando la persona pida escribir/mejorar textos, corregir la redacción, revisar el tono o redactar microcopy (p. ej. "escribe el texto del estado vacío", "haz este mensaje de error más amable", "revisa el titular de la landing").
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。
Genera un nuevo Registro de Decisión de Arquitectura (ADR) copiando la plantilla de ADR en el siguiente archivo numerado y registrándolo en el índice de decisiones. Úsalo cuando la persona quiera registrar una decisión de arquitectura, crear/añadir un ADR o documentar una elección técnica (p. ej. "escribe un ADR para cambiar la base de datos", "añade un registro de decisión").
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。
Trae a este proyecto las mejoras que la plantilla de origen recibió después de la instanciación — lee .template-origin, calcula el diff del tooling (skills, hooks, scripts, workflows) contra la plantilla actual y propone aplicarlo. Úsalo cuando la persona quiera actualizar/sincronizar el proyecto con la plantilla (p. ej. "actualiza la plantilla", "trae las mejoras nuevas del template").
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。
VAE/TAE 解码段优化（**计算 + 通信同技能**）：判断解码器能否沿某个轴切到多张卡、怎么切才是逐位精确、切了值不值；给出可复用的等价性证明流程、状态携带切分代码模式与交换预算判据。范围**含 VAE encode（待补：当前无内容，暂不纳入正文）**。当用户问“VAE/TAE 解码能不能分卡/切分/分摊”“8 张卡重复解码能不能省”“按帧轴还是空间轴切”“切完结果会不会变”“怎么证明等价”“all-gather 拼回来对不对”，或任何“多卡把解码摊开、又要保证输出不变”的场合都应触发；即使用户没说“分片”二字，只要出现“解码每卡都解一遍”“解码太慢想摊到多卡”“那部分重复计算能不能省”也应触发。覆盖：跨轴耦合算子审计（含递归 MemBlock 前缀依赖的否证）、潜帧边界切分＋每块末帧状态携带、与 rank 无关的判定（否则死锁）、交换预算核算、已知静默陷阱（all_gather_into_tensor 填栈副本、repeat_interleave 多分配整张量、rank 选举式参与）、有损分片的精确性前置条件与自动修复、CPU 真值→分段 bitwise→端到端 md5→帧级健康指标的四道验证闸、分片“到底有没有生效”的状态量判定，以及“切分顺带绕过设备缺陷”的用法。边界（本技能只引用不复制）：计算手段与能力供给见 `pattern-dev` / `operator-dev`，DiT 侧选档参照 `dit-perf-opt`；通信方法论（形态/掩盖/通信账/归因）见 `dit-parallel-opt`；一致性判据见 `accuracy-gate`；数字入库见 `perf-gate`；交付搬运与装载预热等 host 固定开销见 `host-opt`（不属本技能）。
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。
Assesses the quality and completeness of a genome annotation with BUSCO (conserved single-copy ortholog recovery), OMArk (proteome completeness, consistency, and contamination), CheckM2 (prokaryotic completeness/contamination), and a gene-set sanity panel (gene count, mono-exonic fraction, protein-length distribution, mRNA:gene ratio, coding density). Covers the assembly-BUSCO-vs-proteome-BUSCO diagnostic, what BUSCO-Duplicated really means, why gene count is a vanity metric, and the QC of transferred annotations. Use when judging whether an annotation is good enough to publish or submit, diagnosing a suspect annotation, or comparing annotation completeness across pipelines.
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。
Stores and operates on sparse expression matrices for single-cell and large bulk RNA-seq, covering dgCMatrix/dgRMatrix/dgTMatrix when-each-is-fast, the dgCMatrix (CSC, R) <-> CSR (Python) implicit transpose, AnnData (cells-rows) <-> SingleCellExperiment (cells-cols) orientation flip, HDF5/h5ad vs Zarr cloud-native shift, HDF5SummarizedExperiment + DelayedArray for out-of-memory bulk, scanpy backed mode for large h5ad, the ~10-15% density crossover where dense beats sparse, 10X format proliferation (MTX vs CellRanger H5 vs h5ad), the dense-conversion memory blow-up, and Dask + Zarr for consortium-scale matrices. Use when choosing sparse format, working with single-cell-sized matrices, importing/exporting 10X, debugging R/Python interop transposes, processing matrices too large for RAM, or building cloud-native pipelines.
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, inspects, and writes Flow Cytometry Standard (FCS) files from conventional, spectral, and mass cytometry (CyTOF), and parses FlowJo/Cytobank/Diva workspaces. Covers FCS 2.0/3.0/3.1/3.2 internals ($PnE linear-vs-log, $DATATYPE, $SPILLOVER vs SPILL vs $COMP, $TIMESTEP), channel/parameter metadata, the silent linearize/truncate defaults, and R (flowCore, flowWorkspace, CytoML) plus Python (FlowKit, readfcs) readers. Use when loading flow or mass cytometry data, mapping detector channels to antibodies, extracting the event matrix, choosing a reader, or bridging FCS to the scanpy/AnnData ecosystem before 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。
Imports gene expression count matrices from featureCounts, HTSeq, STAR ReadsPerGene, Salmon/kallisto via tximport or tximeta, RSEM, 10X Genomics MTX/H5, AnnData H5AD, and RDS. Handles silent-miscounting traps (featureCounts -p v2.0.2 API break, STAR strandedness column choice, salmon NumReads-sum without tximport, RSEM non-integer expected_count, GENCODE _PAR_Y suffix, zero-length-transcript TPM divide-by-zero), and encodes the tximport countsFromAbundance decision tree with the "lengthScaledTPM is not TPM" warning. Use when assembling a gene-by-sample count matrix from aligner or quantifier output, importing salmon/kallisto for DESeq2 vs limma-voom, choosing strandedness column for STAR, debugging zero-count panics, or building tx2gene mapping.
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。
Detects and removes doublets/aggregates from flow, spectral, and mass cytometry before clustering or quantification. Covers FSC-A vs FSC-H singlet discrimination (the Area-Height non-proportionality, not a 1D area gate), FSC-W/SSC width gating, CyTOF Gaussian discrimination parameters (Center/Offset/Width/Residual/Event_length) and DNA intercalator gating, and the residual heterotypic conjugates that survive scatter gating and masquerade as double-positive populations. Use when filtering aggregates before phenotyping, choosing a doublet method for flow vs CyTOF, or diagnosing a suspicious double-positive cluster.
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。