4,120 件 ・ 関連度順
概要と使いどころ
Identifies when task outputs require iteration based on quality signals, unmet requirements, or explicit feedback. Triggers appropriate re-execution strategies. Activate on 'needs iteration', 'retry needed', 'not good enough', 'try again', 'refine output'. NOT for feedback generation (use dag-feedback-synthesizer) or convergence tracking (use dag-convergence-monitor).
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。
Assigns confidence scores to agent outputs based on multiple factors including source quality, consistency, and reasoning depth. Produces calibrated confidence estimates. Activate on 'confidence score', 'how confident', 'certainty level', 'output confidence', 'reliability score'. NOT for validation (use dag-output-validator) or hallucination detection (use dag-hallucination-detector).
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 agent outputs against expected schemas and quality criteria. Ensures outputs meet structural requirements and content standards. Activate on 'validate output', 'output validation', 'schema validation', 'check output', 'output quality'. NOT for confidence scoring (use dag-confidence-scorer) or hallucination detection (use dag-hallucination-detector).
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 fabricated content, false citations, and unverifiable claims in agent outputs. Uses source verification and consistency checking. Activate on 'detect hallucination', 'fact check', 'verify claims', 'check accuracy', 'find fabrications'. NOT for validation (use dag-output-validator) or confidence scoring (use dag-confidence-scorer).
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。
Describe patterns that may indicate bad-faith or manipulative communication including gish gallop, whataboutism, tone policing, motte-and-bailey, sealioning, lying by omission, deflection, and strategic evasion. Use when the user suspects manipulation, notices repeated non-engagement, wants to analyze debate tactics, or mentions 'bad faith', 'deflection', 'manipulation', 'dodging the question', 'avoiding the issue', 'gaslighting', 'moving goalposts'. Describes behavioral patterns without making psychological diagnoses.
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 production-ready LLM applications, advanced RAG systems, and intelligent agents. Implements vector search, multimodal AI, agent orchestration, and enterprise AI integrations. Use PROACTIVELY for LLM features, chatbots, AI agents, or AI-powered applications.
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。
Architecture and systems design for building always-on AI agents with episodic memory. Covers the memory hierarchy (core/recall/archival), persistence layers, agent server infrastructure, vector stores, and framework selection. Provides concrete deployment patterns for agents that maintain identity and learn across sessions. Activate on: "always-on agent", "persistent agent architecture", "episodic memory system", "agent memory design", "long-running agent", "stateful agent", "agent that remembers", "MemGPT architecture", "Letta deployment", "/always-on-agent-architecture". NOT for: choosing what data to feed the agent (use always-on-agent-inputs), brainstorming applications (use always-on-agent-applications), safety and privacy concerns (use always-on-agent-safety), general agentic patterns (use agentic-patterns).
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。
Fast parallel codebase scouting — searches many directories at once, returning a file map + architecture fingerprint. Use before planning or cross-dir work. NOT for deep single-file reads.
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 plan directory (plan.md + phase-*.md) as a visual: generates/validates visual-plan/plan.json, opens a studio to view/edit, exports plan.html. Pass --static for a legacy single-file render.
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。
Serve a markdown file/directory as a calm, book-like reader via a local HTTP server (127.0.0.1 only); good for long-form docs like RFCs and plans. Supports Mermaid, dark/light mode, and a nav sidebar.
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 a plan directory (plan.md + phase-*.md) needs a visual review. By DEFAULT this drives the STRUCTURED visual-plan artifact (visual-plan/plan.json) via the mewkit visual-plan CLI — generate if absent, validate, open the local studio (view/edit), and export plan.html from the approved artifact. Pass --static for the legacy prompt-only single-file plan.html template render. Triggers on "render this plan", "visualize the plan", "open the plan studio", "make a shareable plan page", "plan.html". This is the canonical owner of plan-as-visual-review. NOT for plan critique or scope review (see mk:plan-ceo-review / mk:validate-plan), NOT for generic code/diagram/diff visuals (see mk:preview), NOT for live media generation (see mk:multimodal).
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 publish-grade SVG+PNG technical diagrams — architecture, data flow, flowchart, sequence, agent-memory, class, ER, state-machine, comparison, timeline, and more — across 8 visual styles. Writes output to tasks/visuals/ or an active plan's visuals/ directory. Requires rsvg-convert (librsvg) for PNG export; degrades gracefully to SVG-only when absent.
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 the right agent and complexity tier at Phase 0 orient; runs first in every workflow. NOT for runtime agent routing inside a single skill.
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 guardrails warning before rm -rf, DROP TABLE, force-push, git reset --hard, kubectl delete, etc; user can override. NOT for scoping edits to a directory (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。
Use when creating files and need the correct path, organizing project layout, or enforcing naming conventions. Two modes: advisory (return path) and organize (restructure).
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。
Serve any markdown file or directory as a calm, book-like reader via a local HTTP server (127.0.0.1 only). Best for long-form documents — RFCs, runbooks, design docs, plan files — where comfortable in-browser reading matters. Supports Mermaid diagrams, dark/light mode, plan navigation sidebar, and keyboard shortcuts. Requires explicit start and stop lifecycle.
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 task agent and complexity tier at Phase 0 orient. Invoked first in every workflow. Provider adapters select any model/runtime policy. NOT for runtime agent routing inside a single skill (see orchestration-rules.md).
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 a project has no docs/ directory or needs initial documentation generated from codebase analysis. Triggers on "init docs", "create documentation", "generate docs", "docs init", or when docs/ is empty. Do NOT use for updating existing docs (use mk:document-release).
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。
Frontend development guidelines for React/TypeScript applications. Modern patterns including Suspense, lazy loading, useSuspenseQuery, file organization with features directory, MUI v7 styling, TanStack Router, performance optimization, and TypeScript best practices. Use when creating components, pages, features, fetching data, styling, routing, or working with frontend code.
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 task trajectories to propose reusable SKILL.md candidates from successful patterns
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。
Quantitative agent evaluation using 4-metric framework (correctness/step_ratio/tool_call_ratio/latency_ratio) with ideal trajectory annotation and capability-categorical taxonomy. Use when measuring agent efficiency, comparing agent variants, or gating new agents through correctness→efficiency phases. Complements harness-eval (SE benchmarks) and evaluator-optimizer (qualitative rubric).
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。
Store and semantically search agent memory with ONNX vector embeddings on a ruOS desktop, using the Ruflo CLI. Use to persist patterns/decisions across sessions and recall them by meaning rather than exact key.
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。