「how to」の検索結果

Decides HOW to mutate a DAG when a node fails, quality is below threshold, or new information changes the plan. Selects from mutation strategies (add node, replace agent, fork paths, loop back, downgrade model) based on failure type, cost budget, and execution history. Activate on "DAG failed how to fix", "mutation strategy", "replan on failure", "adaptive DAG", "recovery strategy", "what to do when node fails". NOT for detecting failures (use dag-quality), executing mutations (use dag-planner), or runtime execution (use dag-runtime).

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 and adopt production AI agent infrastructure in 2026. Covers framework selection (LangGraph, CrewAI, AutoGen, MCP), orchestration patterns, evaluation, observability, memory systems, and tool use. Also covers the SOCIAL dimension: how to sell agent infrastructure internally, change management, measuring ROI, building trust in autonomous systems, and scaling adoption across teams. Activate on: "agent infrastructure", "agent framework comparison", "which agent framework", "sell AI tools internally", "agent adoption", "agent observability", "agent evaluation", "MCP architecture", "agentic mesh", "enterprise AI agents", "AI change management", "agent ROI". NOT for: building specific agents (use ai-engineer), designing agent behavior patterns (use agentic-patterns), prompt tuning (use prompt-engineer).

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 a self-contained polished HTML showcase page for a topic/artifacts; mission + paths given explicitly, never auto-scans repo. Writes to assets/showcase/<mission>/. Screenshots need Puppeteer.

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。

Call germline SNPs and indels with GATK HaplotypeCaller and the GVCF joint-genotyping workflow. Covers the local-reassembly + PairHMM mechanism (why HC beats pileup callers on indels), the -ERC GVCF reference-confidence model and <NON_REF> allele, BQSR-vs-DRAGSTR and --dragen-mode error modeling, allele-specific (AS_) annotations, and edge cases (ploidy, Mutect2 mitochondria mode, sex chromosomes/PAR, contamination gating). Use when deciding whether to use HaplotypeCaller vs a pileup or DRAGEN caller, whether BQSR still earns its place, whether to call per-sample GVCFs for a cohort, or how to handle non-diploid, mitochondrial, sex-chromosome, or contaminated samples. Not for post-calling filtering depth (see variant-calling/filtering-best-practices) or cohort joint-genotyping scaling (see variant-calling/joint-calling).

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。

Cycle-end retrospective converting recurring judgment failures into durable enforcement. Reviews the cycle's friction — repeated user corrections, guard trips, RCA themes, integrate failures — and turns each recurring one into a landmines.md entry, then proposes graduation candidates up the enforcement funnel (landmine → advisory hook → hard gate). Pairs with standup: standup says what shipped and what's next; retrospective says what kept going wrong and how to stop it structurally. Writes only memory entries and a report — graduation proposals are advisory and require the normal seed.md amendment path to become enforcement. Invoke at the end of a release cycle, an epic, or any stretch of work with repeated friction.

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。

The primary tool for code-navigation questions in ANY language (frontend or backend) — walk the structural graph from an entry point (page, route, handler, command, `main`) down through imports to the IO boundary (network call, query, file read). Use this skill, in preference to the Explore agent or global grep, whenever the user asks "where does X come from?", "what API/data populates Y?", "what renders or wraps the Z panel?", "which file for feature F?", "how is this wired up?", or any code-navigation question mapping a UI element, route, or feature concept to source files. Also reverse queries ("what uses /api/foo?", "who consumes this service?"). Keyword search routinely picks up unrelated flows sharing a domain word and produces wrong answers; the walk follows the actual graph. The JS/TS `walk.mjs`/`discover.mjs` are an optional accelerator, not required — the walk is language-agnostic.

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。

How to write JSDoc on @rhinestone/sdk public symbols, which is autogenerated into the published SDK Reference. Use when adding or editing doc comments on entry points, the account instance, actions, or utils in sdk/src.

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。

Splits scarce time, budget or traffic across competing products, channels, campaigns or variants by treating them as a multi-armed bandit - Thompson sampling over beta posteriors, with optimism for anything not yet tried enough to judge. Use when deciding what to work on next, how to divide a marketing budget, which product to prioritise, whether to kill something that is underperforming, or when picking a winner among test variants.

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 the ≤10-minute Loom walkthrough script for a completed Weekly Community Signal report — a plain spoken briefing that dives straight into what the team needs to know, walking the report top to bottom (CopilotKit page first, then AG-UI in the same order). Straightforward and factual, NOT a radio show or a performance; no hype opener, no blame framing. Runs after every report (invoked by weekly-report) and on "loom script", "walkthrough script", "record the loom", "narrate the 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。

Use when the user asks about GitNexus itself — available tools, how to query the knowledge graph, MCP resources, graph schema, or workflow reference. Examples: "What GitNexus tools are available?", "How do I use GitNexus?"

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。

Fetch dependency source code to give AI agents deeper implementation context. Use when the agent needs to understand how a library works internally, read source code for a package, fetch implementation details for a dependency, or explore how an npm/PyPI/crates.io package is built. Triggers include "fetch source for", "read the source of", "how does X work internally", "get the implementation of", "opensrc path", or any task requiring access to dependency source code beyond types and docs.

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。

Steward public stories about any work — creation ethics, application ethics, and maintenance of the story plus its disclosures (generated, drafted, curated, authored). Use when writing a public article, origin story, talk, README narrative, or showcase about how work was made; when publishing ethics and decisions; or when asked who generated vs curated a text. Do not use for ADRs, FAQs, skill citations, or marketing hype.

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。

Preview a signed-off design one release phase at a time — cut it down into a series of phase screens, each showing only what exists by that phase with the layout untouched, then put them on a scrubber you drag to watch the screen fill in release by release. Use when the user wants a Phase 1 / MVP version of an existing wireframe or mockup, phased or staged screens, a design sliced by release phase or story map, or wants to see what a screen looks like before later features land.

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。

Produces a deep, honest peer-review of how someone uses Claude Code by statistically analyzing their local session data (~/.claude/usage-data/ and ~/.claude/projects/). Trigger whenever the user asks to review, analyze, audit, or critique their own Claude Code usage, workflow, habits, or skill level — including phrases like "analyze my cc usage", "review my cc sessions", "peer review my cc workflow", "deeper than /insights", or any ask for an honest audit of their AI workflow. Also covers portfolio/hiring-manager framings (e.g. "portfolio for Anthropic/OpenAI/xAI"), but ALWAYS ask the user in Step 0 which version to build (self / hr / both) before running — never silently produce an HR report, because that version goes to outsiders and needs explicit privacy setup. The HTML report is laid out ledger-first for SELF (opening band, output/team/leak/trend ledgers, then peer review, 9-dim scoring, try-this-week, case study, claim-indexed evidence). The HR variant is a recruiter version v1: identity card, earned badges (threshold-based, earned-only), allowlist-filtered output ledger, one case study, and a methodology & scope disclosure — no scores, no peer-review memo, no charts.

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 editorial/magazine/report webpages on a GENUINE Müller-Brockmann modular grid (International Typographic Style). Encodes the discipline (columns + modules + baseline, grotesque type, flush-left, restrained black/white/red palette) AND the front-end engineering to make the grid real, visible, and verified: one CSS-variable source of truth, an interactive grid-toggle overlay, subgrid bands, 8px baseline lock, and runtime optical alignment. Triggers: "magazine spread", "grid system", "Swiss design", "editorial layout", "show the grid", "align everything to the grid".

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。

Attach a self-contained Excalidraw status diagram to a GitHub epic — read the epic + its sub-issue tree, render boxes-coloured-by-status with dependency arrows to a committed PNG that (a) renders on the GitHub mobile app and (b) has the editable Excalidraw scene embedded inside it, post it as a sticky comment, and round-trip human edits back in from a comment. Use when asked to "diagram an epic", "draw the ticket tree", "show the epic on mobile", "add a status graph to

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。

Render a crouton collection schema (the field-definition JSON) into a human-readable field table + relationships sketch (HTML + PNG + a terse Markdown table), so a human can sign off on the DATA MODEL before any code is generated. Use when an agent drafts or changes a collection schema (before `crouton config` / `generate_collection`), or when asked to "review a schema", "show the data model", "check these fields".

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。

System reference for the crouton layout engine (@fyit/crouton-layout) — the LayoutTree/layout_configs data model, the croutonLayoutBlocks registry and sizing contract, the viability metric, the deterministic placer (#709), and how layouts persist and seed. Use when reasoning about WHY a layout renders/composes/persists the way it does, when a question is shaped like "what is a LayoutTree / where do layouts live / why did the placer pick this arrangement / why won't this pane shrink / where does minWidth get read", or when a generated POC boots with the wrong (or no) default layout. Trigger phrases: "layout tree", "layout_configs", "viability check", "composeDefaultLayout", "block sizing contract", "why is my layout not viable". For authoring a block component, use the block-authoring skill 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。

Day-to-day operation runbook for nuxt-crouton apps — boot any app/POC locally with auth working, seed and inspect the local database (including the miniflare-vs-.data split that makes seeded data "disappear"), turn on the review overlay / eruda, observe deployed staging (tail, smoke, row counts), and capture screenshots. Use when asked to "run the app", "boot velo/fanfare/a poc locally", "why can't I log in locally", "seed the database", "the seed data doesn't show up", "look at the local db", "check what's on staging", "screenshot the app", or "where do the artifacts land".

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。

Plan, run, and measure an adversary engagement operation using the MITRE Engage framework so that deployed deception is driven by strategy instead of deployed ad hoc. Covers the Engage Matrix (Prepare, Expose, Affect, Elicit, Understand), the 10-Step Operational Process, mapping engagement Activities to the ATT&CK techniques they expose, and defining measurable Goals and Operational Objectives. Use when a team has honeypots, honeytokens, or canary tokens but no coordinating strategy, when leadership asks "should we engage attackers and how", when building a deception/denial program, when writing an adversary engagement operation plan, or when deciding which deception Activities to deploy against a specific threat actor. Keywords: MITRE Engage, adversary engagement, cyber deception strategy, denial and deception, Engage Matrix, EAC, EGO, Expose Affect Elicit, deception program, honeypot strategy, engagement operation.

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。

Show what maintenance robots already did today before you repair anything by hand: backups, reindexes, guards, quotas and what went stale. Use when deciding whether to rerun a backup, or when a command was blocked by a once-a-day gate. Triggers: /hk, /housekeeping, what was done today, should I run the backup, why was this blocked, show quotas, housekeeping status.

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。

Get external links to your repos through other people's curated surfaces: awesome-lists, tool catalogs and vendor showcases. Measures each list's merge queue before submitting, so nothing goes into a dead queue. Triggers: /borrowed-audience, /ba, borrowed audience, get external links, submit to awesome lists.

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。

Analyze the current repo structure, build system, test setup, and conventions to provide a practical onboarding guide. Use when new to a codebase, joining a project, or wanting to understand how a repo is organized. Triggers on getting started, new to repo, onboard, how does this repo work, repo structure, codebase overview.

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。

Interactive guided tour of all available AI coding skills with live demos. Walks through headline capabilities, offers try-it-now demos, discovers repo-specific tools, and provides a cheat sheet reference. Triggers on what can you do, show skills, skill tour, available tools, capabilities, what 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。