4,696 件 ・ 関連度順
概要と使いどころ
Review all uncommitted work and commits not yet pushed on the current branch, fix credible findings, commit remaining changes, and push. Use when the user invokes `$push` or asks to review, commit, and push all current branch 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。
Validate trained models and analytical claims against their intended decision, independent evidence, and human-reviewed ground truth. Use when reviewing model performance, analysis conclusions, launch claims, or evaluation reports; use failure diagnosis instead when the main task is locating a known defect.
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。
Coordinate complex engineering work across bounded subagents while protecting the main context, preventing edit conflicts, and measuring whether delegation was worthwhile. Use when the user asks to delegate, parallelize, orchestrate, use subagents, fan out a review, or handle a large task with independent workstreams; do not use for small, tightly coupled, or inherently sequential 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。
Turns a vague "what should we do about X" into a scored decision - explicit actions, an explicit unknown, an explicit prior, and one utility scale - then picks by maximum expected utility and checks the result for framing effects. Use when facing a choice between options under uncertainty, when a decision has stalled in circular debate, when someone asks "should we do A or B", or before writing any decision record, strategy memo, or roadmap commitment.
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。
Models what a competitor will do next and what happens if you respond - best response, equilibrium, and the difference between a rival who optimises perfectly and one who does not. Use before a price change or price war, when a competitor launches something, when deciding whether to match a rival's move, when analysing a market with few players, or when someone asks "what will they do if we do this".
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。
Decides how far ahead to plan, what to discount future payoffs by, and which backlog items can be dropped without analysis - using receding-horizon planning, an explicit discount factor, and branch-and-bound pruning against the incumbent. Use when building a roadmap, prioritising a backlog, arguing about short-term versus long-term, setting quarterly or annual plans, or when planning has become an end in itself.
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。
Validates a plan before committing to it - checks whether the ranking survives the assumptions, finds the most likely way it fails, and makes the trade-offs explicit via a Pareto frontier instead of an invented single score. Use before a launch, price change, migration, infrastructure change or major commitment, when running a premortem, when two objectives conflict, or when a decision depends on an assumption nobody has tested.
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 past decisions into calibrated judgment - scoring old predictions against what happened, assigning credit for delayed results, and separating a bad decision from a good decision that lost. Use during a retrospective, postmortem, quarterly or monthly review, when reviewing decisions taken 30+ days ago, when attributing a result to a cause, or when someone claims to have called something.
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。
Prices research, tests, surveys, dashboards and experiments before running them, by computing how much the result would raise expected utility - which is zero whenever no outcome would change the decision. Use when someone proposes an A/B test, user survey, market study, competitor analysis, analytics build, consultant, pilot, or "let's gather more data", and when deciding how large a study needs to be.
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。
Separates signal from noise in metrics you cannot observe directly, updating a belief from evidence instead of reacting to the latest reading - Bayesian updates over competing explanations, and a filter that says whether this week's move is real. Use when a metric moves and someone wants to act, when diagnosing why traffic or revenue changed, when a dashboard number contradicts intuition, or when deciding whether a trend is real yet.
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 and finalize one ready rolling-wave slice in one invocation while preserving the lifecycle boundary between implementation and independent review. Use when the user wants a slice delivered end to end without manually invoking implement-slice and finalize-slice, wants main-local implementation by default with workers only when delegation pays, followed by mandatory fresh-context finalization, or says deliver, implement and finalize, or take the slice through review.
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。
Finalize an implemented rolling-wave slice for user review. Use when the user wants Codex to review the current branch state against a ready-for-review or in-review slice contract, confirm and fix required findings, repeat review and verification until no required findings remain, ask before accepting findings that materially change project shape or contradict project assumptions, record review state, and leave the clean slice in review for complete-slice.
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。
Shared support references for the rolling-wave engineering skill set. Use only when a rolling-wave workflow skill explicitly points here, or when maintaining the shape-project, prepare-next-slice, implement-slice, finalize-slice, and complete-slice 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。
Implement a ready rolling-wave slice without locking it down with broad tests yet. Use when the user explicitly wants to code the current docs/rolling-wave/{project}/slices/NNN-slug.md slice, mark it in progress while working, implement small or cohesive work in the main context by default, use one to three gpt-5.6-terra workers only for genuine parallel chunks or large discovery/mechanical work where delegation pays for its handoff cost, pause or hand off chunks marked human-fit when appropriate, record implementation notes and final-test backlog, preserve the original slice contract, and mark completed implementation ready for review.
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。
Prepare and risk-check the next rolling-wave slice for implementation. Use when the user wants to continue a rolling-wave project, draft one bounded slice in the main context by default, use a fresh planning author only when context or discovery cost warrants it, use a fresh preparation reviewer when semantic risk or uncertainty warrants independent pressure-testing, decompose parallel implementation chunks where appropriate, identify agent, human, either, or hybrid execution fit, write or update docs/rolling-wave/{project}/slices/NNN-slug.md, and mark that slice ready even when other slices are active.
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 repository documentation authority, repair reading paths and explanations, or shape a bounded public projection. Maintain docs after a bounded change when indirect reader promises, generated or paired sources, version/evidence boundaries, or unclear authority need attention. Ordinary known updates stay in normal development; not for isolated prose polishing, code review, or broad runtime audits.
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 a comprehensive code quality audit covering security, performance, and maintainability. Use for quarterly audits, pre-release reviews, or when onboarding to a large codebase.
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。
Find overly complex code: files over 300 lines, functions over 50 lines, deeply nested conditions (4+ levels), and magic numbers. Use for code quality reviews, pre-release checks, or when asked to assess code 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。
Subagent flow for deep-reading GitHub issues + their linked/closing PRs to the bottom. Captures CURRENT state, maintainer status updates, reporter repro, blocking reviewer concerns, hidden second bugs, cross-repo duplicates. Invoked by weekly-report orchestrator; runs as a general-purpose subagent so the orchestrator's context stays small.
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-level control systems engineering covering feedback control, PID design, root locus, frequency domain methods, state-space design, and digital control.
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-level clinical engineering covering medical equipment management, hospital technology systems, healthcare IT integration, safety standards, and health technology assessment.
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-level inorganic chemistry knowledge. Use when working with coordination chemistry, transition metals, crystal field theory, organometallics, main group chemistry, solid state chemistry, acid-base theory, or inorganic reaction mechanisms. Also use when the user mentions 'coordination complex', 'ligand', 'crystal field', 'transition metal', 'oxidation state', 'organometallic', 'Lewis acid', 'VSEPR', 'symmetry', 'point group', 'solid state', or 'catalysis'.
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-level number theory knowledge. Use when working with divisibility, prime numbers, modular arithmetic, congruences, Diophantine equations, cryptography, quadratic residues, or analytic number theory. Also use when the user mentions 'prime', 'divisibility', 'modular arithmetic', 'congruence', 'GCD', 'Euler totient', 'Fermat little theorem', 'Chinese remainder theorem', 'quadratic residue', 'Diophantine equation', 'RSA', or 'prime factorization'.
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。