495 件 ・ 関連度順
概要と使いどころ
AI specification planning frameworks. Use when a spec touches model calls, prompts, retrieval, tool calling, agentic loops, or evals. Covers approach selection, model and provider choice, structured output contracts, loop guards, budgets, failure modes, and eval design.
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。
Provider-agnostic patterns for LLM function calling, tool loops, and agentic workflows
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。
Phase 4 of the feature-pipeline suite. Use this after /fp-review has posted a sandbox review comment — it reads the outstanding checkbox items, applies fixes, then watches CI until green (or exits cleanly so you can retry). Trigger phrases: "address review items", "fix review feedback", "resolve review checklist", "watch CI", "poll CI", "check if CI is green", "iterate on PR", "fp-iterate", "Phase 4", "after review comment", "fix the review issues and watch CI". Always bounded — max 10 poll rounds, never loops forever. Exits with a clear status: green (hands off to /fp-drift), still-pending (re-invoke), or failed (diagnose before touching anything).
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 position-independent code and shellcode by identifying GetPC/PEB-walk techniques, API-hash resolution loops, and relocation-free addressing so the payload can be disassembled at the right base. Activates for requests to analyze position-independent code, identify PIC techniques, or understand how shellcode resolves APIs without imports.
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 building or reviewing Canvas 2D, Three.js/WebGL, React Three Fiber, GLSL shaders, ShaderToy-to-WebGL adaptation, 2D/3D visualization, game rendering, animation loops, GPU resource cleanup, or rendering performance. Do not use for standard DOM UI, charts already covered by a chart library, or non-graphical performance work; Chinese triggers include Canvas, Three.js, WebGL, GLSL, Shader, 3D, data visualization, games.
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 behavior-preserving frontend refactoring focuses on behavior/data ownership, statement placement, loops, or dead-code knowledge, and the selected technique needs family-specific mechanics, risks, and JavaScript/TypeScript/React/Vue adaptation.
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。
Iterative manuscript improvement workflow where a reviewer agent critiques, an author agent revises, a verifier checks the revision, and the loop repeats until acceptance or a human-pause condition is reached. Use for virtual peer review, reviewer-author revision loops, rebuttal preparation, resubmission polishing, journal paper improvement, response-to-reviewers drafting, and deciding when additional data, experiments, theory, modeling, or author judgment is required before further revision.
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。
Execute ENCODE Hi-C pipeline from FASTQ to contact matrices and loop calls. Child of pipeline-guide. Provides Nextflow execution with Docker and cloud deployment. Use when processing Hi-C data, generating contact matrices, or calling loops. Trigger on: Hi-C pipeline, chromatin conformation, contact matrix, loop calling, TAD detection, Juicer, HiCCUPS, 3D genome.
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 comprehensive chromatin contact maps by aggregating Hi-C loop calls (BEDPE) across multiple ENCODE experiments, donors, and labs. Use when the user wants to answer "what regions are in 3D contact in my tissue?" by creating a union catalog of chromatin loops. Handles resolution-aware anchor matching, cross-lab variation, and Hi-C-specific quality metrics.
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 productive with agentic coding assistants — task framing, iterative prompting, file-scoped workflows, review loops, and multi-file refactors. Use when pairing with an AI coding agent on real codebases.
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 agents that work toward goals with minimal supervision — goal decomposition, self-correction loops, safe autonomy levels, and long-horizon task management. Use when an agent must run for hours or complete open-ended 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。
Run goal-driven autonomous agent loops — decomposing objectives, executing tool chains, self-critiquing, and iterating until done. Use when exploring what fully autonomous task execution looks like and its limits.
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。
Architect autonomous agent systems — perception/action loops, tool design, planning strategies, memory, and multi-agent orchestration patterns. Use when designing an agent from scratch or scaling a prototype to production.
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 and build LLM agents: ReAct loops, tool schemas, planning, memory, guardrails, evals, and human-in-the-loop. Use when building autonomous multi-step AI systems.
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。
Ship AI features to production — model selection, evals, latency/cost optimization, guardrails, monitoring, and iteration loops. Use when turning an LLM prototype into a reliable product feature.
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。
AI agent guidance — agent architecture, tool design, planning loops, guardrails, evaluation, and production agents.
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。
Scaffold an isolated CLOUD dev box per agent (via crabbox + Daytona) for any codebase — the parallel-safe counterpart to dev-local-setup. Each agent gets its own full stack (own DB + dev server) and an in-box browser for e2e, so concurrent loops never collide on ports/state. Sets up the snapshot image, .crabbox.yaml, an idempotent setup.sh (also boots the stack locally), and a cbx.sh wrapper. Use when the user says "set up crabbox", "give each agent its own box", "add cloud testing", "make this repo testable in the cloud / on Daytona", "parallel-test this", or when setup-codebase-harness needs true per-agent isolation.
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, implement, audit, and review C#/.NET SwiftlyS2 plugins. Use when working with Commands, Core Events, Game Events, GameHooks Pre/Post, raw native hooks, Modules, Workers, Services, high-frequency runtime loops, NetMessages, Schema access, entity handling, thread safety, performance, or IPlayer lifecycle behavior.
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 debug OAuth 2.0 DPoP (RFC 9449) refresh token sender-constraining for WebCrypto, Node.js ES6, and browser runtimes integrating with Google's OAuth platform. Use when configuring non-extractable asymmetric key pairs (P-256), generating DPoP Proof JWTs for authorization code exchange and token refresh, or handling 400 use_dpop_nonce challenge retry loops at oauth2.googleapis.com/token. Don't use for unconstrained OAuth 2.0 flows (where refresh tokens are not bound to a client key pair), or for Google Cloud IAM / service account authentication.
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。
Diagnoses GKE JobSet interruptions, restarts, and preemptions for AI/ML training workloads autonomously. Use when troubleshooting JobSet restart loops, spot VM preemptions, node readiness failures, host VM issues, or coordinator worker crashes. Don't use for general GKE cluster creation, basic workload deployment, or non-JobSet application issues.
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 work as an explicit event-driven state machine executed by subagents. The orchestrator is the runtime from the canonical loop — `state, effects = init(machine); execute(effects); while state is active: event = nextEvent(); state, effects = transition(machine, state, event); execute(effects); return state.output`. All routing judgment is authored ONCE, at plan time, into a machine file (states, event alphabet, transition table, budgets, final states); at runtime the orchestrator does pure table lookups and dispatches effects — it never decides, never authors, never ingests artifacts. Subagent/sync protocol is the timeboxed-iterating one: role prompts scaffolded once to disk and dispatched BY PATH, an initialiser preamble every subagent reads first, a self-populating cheatsheet, per-unit files, tiny structured returns, independent measurement gate. Use when the user says "workmachine", "run this as a state machine", "state-machine this", or when a goal has a small number of clearly named phases with explicit outcomes (built / passed / failed / blocked) and you want the routing to be auditable rather than judged on the fly. For clock-driven loops use timeboxed-iterating; for wide graphs with judged edge criteria use workgraph.
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。
Orchestrate AI coding agents (Claude Code, Codex, OpenCode) in isolated sandboxes using the @ai-hero/sandcastle SDK. Use when the user needs to (1) run agents AFK in Docker/Podman containers, (2) build multi-agent pipelines with plan-execute-review patterns, (3) run parallel agents on separate worktrees, (4) create iterative agent loops with maxIterations, (5) extract structured output from agent runs, (6) set up sandcastle in a new or existing project, or (7) write prompt files with template args and shell expressions.
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。
Ideate a brand-new game concept with the user, from a blank page or a one-line hint to a finished CONCEPT card. Use it whenever the user wants to come up with a new game — whether the project has no design document yet, or already has one and they're exploring a fresh idea. Through a structured divergence/convergence dialogue it explores ideas, frames the hook, generates and scores concepts (market validation delegated to /unikit-gd-explore), and settles pillars, loops, motivation, and a pre-mortem. Trigger whenever the user wants to make or invent a game, doesn't know where to start, or gives a genre/theme hint, e.g. "let's come up with a game", "I don't know what game to make", "I want to make a roguelike", "a game about zombies", "a farming game", "help me come up with a concept". For researching or dissecting existing games use /unikit-gd-explore; to write an already-chosen concept as the master GDD use /unikit-gd-spec.
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。
Autonomous comprehensive refactoring. Three-phase pipeline — Phase 1 /refactor (tactical, loops internally to convergence) → Phase 2 loop of /review-arch + /implement-batch until architectural findings converge below a severity floor (default HIGH+) → Phase 3 /refactor (catches tactical issues introduced by restructuring). Auto-approves /review-arch ticket proposals; pulls the andon cord on contested findings, breaking-change proposals, repeated implementation failure, or Phase-2 hard-cap exhaustion (5 architectural iterations).
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。