「each」の検索結果

Fundamental patterns for effective agentic behavior. Teaches decomposition, tool orchestration, error recovery, context management, quality self-assessment, and knowing when to stop. Model-agnostic principles that make any agent more effective regardless of domain. Activate on: "how should I structure this agent", "agentic workflow", "agent patterns", "multi-step task", "tool orchestration", "/agentic-patterns", "decompose this", "agent best practices", "chain of actions", "when should the agent stop", "agent loop design". NOT for: creating agent infrastructure (use agent-creator), building DAGs (use windags-architect), specific tool implementation.

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 reviewer comments on a GitHub PR: verifies each against the codebase, then accepts/pushes back/clarifies, replies in-thread. NOT for reviewing (mk:review-pr); NOT for fixes (mk:fix).

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。

Resolves an in-progress git merge/rebase conflict: reads conflict state, recovers each side's intent, resolves hunks, runs checks. NOT for the ship pipeline (mk:ship); NOT for bug debugging (mk:fix).

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。

Systematically QA-tests a web app and fixes bugs: test, then iteratively fix, re-verify, commit each fix. Tiers: Quick, Standard, Exhaustive. Produces before/after health scores and fix 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。

Coordinates staged or parallel jobs across live-verified coding-agent runtimes (Claude Code, Codex, Cursor, and other installed CLIs) and in-session toolkit subagents. Routes each job by capability and risk tier, isolates parallel writers in git worktrees, captures redacted output, resumes interrupted runs, and blocks completion until an independent arbiter verifies the results. Use when work should be split across multiple runtimes or subagents, run as staged or parallel jobs, and reviewed before handoff. NOT for a single-agent task or the in-session 7-phase flow (use mk:cook / mk:workflow-orchestrator).

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。

Triage reviewer comments on a GitHub PR with receiving-code-review discipline — verify each against the codebase, then accept/push-back/clarify, and optionally reply in-thread. Use when responding to received PR feedback. NOT for giving a review (see mk:review-pr); NOT for implementing the fixes (hand off to mk:fix).

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。

Systematically QA test a web application and fix bugs found. Runs QA testing, then iteratively fixes bugs in source code, committing each fix atomically and re-verifying. Use when asked to "qa", "QA", "test this site", "find bugs", "test and fix", or "fix what's broken". Proactively suggest when the user says a feature is ready for testing or asks "does this work?". Three tiers: Quick (critical/high only), Standard (+ medium), Exhaustive (+ cosmetic). Produces before/after health scores, fix evidence, and a ship-readiness summary.

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。

Resolve an in-progress git merge or rebase conflict end to end: read the conflict state, recover each side's original intent, resolve every hunk, run the project's checks, and prepare the merge/rebase for an explicitly approved finish. Use when asked to "resolve merge conflicts", "fix this rebase", "finish the merge", or when `git status` shows conflicted (unmerged) paths. NOT for the full ship pipeline that merges the base branch first (see mk:ship); NOT for debugging a runtime bug or test failure unrelated to a conflict (see mk:investigate / mk:fix).

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。

Session-scoped safety guardrails for destructive commands. Warns before rm -rf, DROP TABLE, force-push, git reset --hard, kubectl delete, and similar destructive operations. User can override each warning. Active for the current session only. Use when touching prod, debugging live systems, or working in a shared environment. Use when asked to "be careful", "safety mode", "prod mode", or "careful mode". NOT for scoping edits to a specific directory (see mk:freeze).

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。

Relentlessly interview the user about THEIR plan or design — one question at a time — resolving each branch of a decision tree until shared understanding. Claude is the interviewer; the user is the interviewee. Triggers on "grill me", "stress-test this plan", "get grilled on my design", "interrogate my plan". Checkpoints every answer to docs/knowledge/<slug>.md so progress survives a context-window reset. Flags knowledge the user cannot answer as a gap to chase. Do NOT use to PROPOSE solutions or compare approaches (→ mk:brainstorming). Do NOT use to validate "is this worth building?" (→ mk:office-hours). Do NOT use when Claude should ANSWER questions about the repo (→ mk:ask-me). Do NOT use for a one-shot review of an existing plan (→ mk:plan-ceo-review). Do NOT use to apply named reasoning lenses to findings (→ mk:elicit).

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。

Turn a raw idea into ONE honest recommendation: interview one question at a time until the problem, requirements, goals, non-goals, and constraints are CONFIRMED, then deliver a single verdict — do, don't, cheaper alternatives, trade-offs, ordered checklist, success metrics. Required: the user wants exactly ONE recommendation — not options, not a debate — and their framing is open to challenge (advise may answer a different question than the one asked). Triggers on "what should I do about X", "advise me on X", "am I approaching this right", "is this the right move". Do NOT use to be interrogated with no recommendation (→ mk:grill) — advise also interviews one question at a time, but only to reach a verdict. Do NOT use when the framing is settled and only the options are open (→ mk:brainstorming). Do NOT use to decide whether something is worth building (→ mk:office-hours). Do NOT use when you want several perspectives to argue it out (→ mk:party).

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 ONLY when explicitly invoked. Puts Fusionaly on a Linux server you reach over SSH with a key. Optionally creates a Hetzner server and the Cloudflare DNS record. The user runs the one interactive install command; the agent prepares and checks everything around 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。

Monitor Claude Code releases and auto-generate GitHub issues for each new version

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。

Multi-LLM adversarial consensus loop — 3+ LLMs compete to find flaws in designs/specs until unanimous agreement is reached

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。

Build visual territories (mood boards) for team alignment before production. Groups references into 2-4 named territories, each with palette, texture, typography feel, and 4-6 generated sample tiles, so a team can agree on a direction. Use when the user says "moodboard", "visual territories", "align the team on a look", "reference board", or needs stakeholder sign-off on aesthetic before generating final assets. Part of the Creative Production set.

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。

Rapidly generate up to 25 distinct ad creative hypotheses for testing. Produces a matrix of angles (pain, benefit, social proof, urgency, curiosity, comparison) x formats (static, carousel hook, UGC-style, bold-text overlay) with headline, visual prompt, and rationale for each, then generates the visuals with any image tool. Use when the user wants "ad variations", "ad concepts", "creative testing batch", "give me 25 ads", or paid-social creative ideation. Part of the Creative Production set.

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。

Reflect subskill for simmer. Records iteration results in trajectory table, tracks best candidate, handles regression rollback, and passes ASI forward to the next round. Supports both single-file and workspace modes. Do not invoke directly — called by simmer orchestrator after each judge round.

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。

Judge board subskill for simmer. Dispatches a panel of judges with different lenses, runs one deliberation round where they challenge each other's scores, then synthesizes consensus scores + single ASI. Drop-in replacement for simmer-judge that produces identical output format. Do not invoke directly — dispatched by the simmer orchestrator when JUDGE_MODE is board.

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。

Analyzes single-cell pooled CRISPR screens (Perturb-seq, CROP-seq, Perturb-CITE-seq, ECCITE-seq, multiome) where each cell carries an sgRNA and a scRNA-seq / surface-protein / chromatin readout. Covers experimental design (direct-capture Perturb-seq Dixit 2016 vs CROP-seq 3'UTR-barcoded Datlinger 2017 vs ECCITE-seq vs Multiome), MOI for sgRNA assignment, escaper-cell filtering (Mixscape, Papalexi 2021), SCEPTRE NB GLM + permutation for low-MOI (Barry 2024 Genome Biol 25:124), the Pertpy framework, factor decomposition, genome-scale Perturb-seq (Replogle 2022 Cell, 2.5M cells), and per-perturbation single-cell DE. Use when running a single-cell CRISPR screen, choosing direct-capture vs CROP-seq architecture, filtering escaper cells, performing single-cell DE, integrating Perturb-seq with pathway analysis, scaling to GW CRISPRi via Replogle protocol, or analyzing multi-omics screens.

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 somatic mutations in circulating tumor DNA, treating low-VAF detection as a signal-versus-noise problem set by error suppression and molecules sampled, not by the choice of caller. Distinguishes de novo CALLING (scanning a panel for unknown variants, bounded by per-locus error and multiple testing) from tumor-informed DETECTION (tracking a pre-specified variant set, where panel integration reaches single-ppm). Covers VarDict and Mutect2 for de novo calling, UMI-aware callers, and a pysam-based known-variant VAF tracker, with matched-WBC subtraction as the mandatory defense against clonal hematopoiesis (the dominant false positive). Use when calling or tracking tumor mutations from plasma cfDNA, setting a VAF threshold, or deciding whether a low-VAF call is tumor versus CHIP.

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。

Cross-method decision tree for calling hits in pooled CRISPR screens. Catalogs statistical models (MAGeCK RRA, MAGeCK MLE, BAGEL2, drugZ, JACKS, Chronos, CERES), experimental designs each is built for, failure modes outside design domain, reconciliation when methods disagree, multiple-testing and effect-size thresholds, the order of operations (count -> QC -> CN-correct -> hit-call -> validate), the second-best-sgRNA conservative rule, and consensus-hit strategy. Use when choosing among MAGeCK / BAGEL2 / drugZ / JACKS / Chronos for a given design, reconciling disagreement across two or three methods on the same screen, deciding whether to require consensus, gating downstream validation by hit-confidence tier, or interpreting unstable hit lists across reruns.

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。

Identifies differentially bound ChIP-seq regions between conditions using DiffBind, csaw (sliding windows), DESeq2/edgeR/PyDESeq2 on count matrices, NormR (control-aware), or MAnorm2. Distinguishes three distinct normalization problems (composition bias, trended bias, global shifts) and matches each to its appropriate fix including spike-in scaling. Use when comparing ChIP-seq binding between experimental conditions, choosing normalization for global vs local changes, integrating spike-in data, or reconciling DiffBind/DESeq2 disagreement.

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。

Treats a ctDNA assay as a molecule-counting experiment at the Poisson edge and builds its analytical-validation case the measurement-science way. Covers the genome-equivalent currency (~330 haploid copies/ng), the lambda = input_GE x VAF sampling ceiling (lambda>=3 for ~95% detection), the error-suppression ladder (raw NGS ~1e-3 -> single-strand UMI ~1e-4/1e-5 -> duplex <1e-7), the CLSI EP17 LoB/LoD/LoD95/LoQ framework, the per-locus-vs-panel-integrated LoD distinction that lets bespoke MRD reach ppm, contrived/SEQC2 reference standards, and honest LoD reporting conditioned on input mass + consensus depth + replicate detection rate. Use when stating or trusting a sensitivity claim, designing a dilution-series validation, deciding how many genome equivalents are needed at a target VAF, choosing a single-locus vs panel-integrated LoD, or auditing a "detects 0.1% VAF" claim.

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。