「dpo」の検索結果

Local LLM inference with the Ollama JavaScript client -- chat, streaming, tool calling, vision, embeddings, structured output, model management, and OpenAI-compatible endpoint

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。

Serverless GPU compute platform for AI model deployment — web endpoints, GPU functions, model serving, and TypeScript client 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。

Hugging Face Inference SDK patterns for TypeScript/Node.js — InferenceClient setup, chat completion, text generation, streaming, embeddings, image generation, audio transcription, translation, summarization, and Inference Endpoints

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 or revise Interpreter UI for non-developers with an ultra-minimal, utilitarian style inspired by the OpenAI website and ChatGPT, not the developer platform. Use when simplifying settings, onboarding, chat, forms, empty states, or helper copy; when removing extra containers, labels, or visual noise; when replacing technical wording with plain language; when clarifying hierarchy and making the main user action obvious; when making every visible label and status understandable to an average HR admin or other non-technical user; when hiding implementation details such as logs, IDs, endpoints, paths, or protocol names; or when a screen must respect the app's existing CSS variables and light/dark theme 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。

Burn timed, styled captions into a finished video on the local machine, without re-rendering the source. This is the manual fallback to the API's captioning endpoint, POST /v1/captions, which stays the default path: come here when the style the user wants is outside the API's presets, when the wording has to be hand-corrected before it is burned into the pixels, or when the source was rendered with audio disabled and the API refuses it. Transcribes the audio with Whisper, groups words into reading phrases, renders the captions in a real browser via HyperFrames, and composites them over the source with ffmpeg using a real alpha channel. Use when the user has an MP4 with speech and asks to add captions, subtitles, burned-in text, TikTok-style captions or word-by-word text to it, and the API path has been ruled out. Runs out of band, so it costs no credits and works on any video file whoever generated it. Not for generating video (use the video skills) and not for a separate .srt/.vtt sidecar file.

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。

Walks a user through connecting the judge models this plugin can use — Codex, OpenCode, OpenRouter (or any compatible endpoint: 9router, z.ai, Moonshot), Gemini — and through picking models, including "research current models for me". Use for "multi setup", "set up multi", "configure judges/reviewers/ backends", "connect 9router", "change/pick models", "onboarding", or whenever a probe shows missing backends and the user needs to know what to do about 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。

Plan a feature before writing any code. Use this skill at the START of non-trivial feature work — when the user says they want to "add", "build", "implement", "create", "develop", or "ship" a feature, endpoint, component, page, service, or capability. Especially trigger when the user's prompt is short or under-specified ("add a dashboard", "build an auth flow", "let's do user activity reports") — those need planning, not direct execution. The skill walks through requirements clarification, codebase exploration, sketches a Mermaid architecture/flow diagram of the proposed design, and produces a plan.md document the user can hand off to a fresh AI session for execution. Skip only when the change is genuinely trivial (typo fixes, single-line tweaks, renames). When in doubt, plan.

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。

Captures and analyzes malware network traffic from a detonation: extracting C2 endpoints, HTTP/DNS patterns, TLS metadata (JA3/SNI), and beaconing from a PCAP to build network IOCs and detection. Activates for requests to analyze malware PCAP, extract C2 from traffic, or study network 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。

Analyzes cryptojacking/coinminer malware by extracting mining pool endpoints (stratum), wallet addresses, algorithm and miner identifiers (XMRig and forks), and persistence or resource-control settings from static strings and config. Activates for requests to analyze a cryptominer, extract mining pool and wallet IOCs, or identify coinminer configuration.

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。

Security-first code review & vulnerability prevention (IDOR, XSS, CSRF, SSRF, SQLi, Auth/JWT, Mass Assignment, Sensitive Data Exposure). Use when writing web features, designing endpoints, or auditing security.

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 frontend work needs to communicate data, action, state, permission, validation, or business-rule needs to backend teams without dictating endpoint design, field names, database shape, or implementation details; Chinese triggers include front-end and back-end requirements handover, back-end requirements, API requirements clarification, data requirements description.

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 task needs the judgment of an Application Security Engineer — checking whether an endpoint verifies object-level authorization (not just authentication), running threat modeling at design time rather than deferring security review to the end, triaging a dependency vulnerability by actual reachability rather than CVSS score alone, combining SAST/DAST/SCA tooling to cover their distinct blind spots, or applying context-specific output encoding to untrusted input. Distinct from an information security analyst — this role works inside the SDLC to prevent vulnerabilities before they ship, not detect and respond to intrusions in 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。

Spring Security 7 implementation for Spring Boot 4. Use when configuring authentication, authorization, OAuth2/JWT resource servers, method security, or CORS/CSRF. Covers the mandatory Lambda DSL migration, SecurityFilterChain patterns, @PreAuthorize, and password encoding. For testing secured endpoints, see spring-boot-testing 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。

Use this when the user wants to generate, normalize, verify, deduplicate, or export training datasets for Codex, Antigravity, or Claude Code from topics, URLs, reference material, web research, or existing JSONL/CSV files. Supports SFT and DPO workflows, custom export schemas, and deterministic local pipeline scripts.

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 and manage Volcengine cloud resources using the Volcengine CLI (`ve` command). Supports all Volcengine services including ECS, VPC, CLB, RDS, Redis, and more. Trigger this skill whenever the user asks to create, query, modify, or delete cloud resources on Volcengine, mentions the `ve` command, says "volcengine CLI", or describes infrastructure tasks such as "create an ECS instance", "set up a VPC", "list security groups", "allocate an EIP". Also trigger on Chinese prompts mentioning "火山引擎" or "火山" (e.g., "火山引擎上有哪些 ECS"、"查一下我火山的云服务器"、 "火山引擎创建一个 VPC"、"火山的 Redis 实例列一下"). Also trigger when the user encounters errors from `ve` commands and needs troubleshooting help. Also use for TOS bucket/object operations, tosutil (sometimes written tosutils), Ark CLI / arkcli model discovery, inference, generation, endpoints, fine-tuning, plans, or usage, and TLS (日志服务) / volclog log search, analysis, ingest, export, or LogCollector collection, through `ve tosutil`, `ve arkcli` and `ve volclog`.

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。

Invoke a Rubber Duck Reviewer subagent to independently critique plans and implementations before proceeding. Use when the agent is about to implement a non-trivial plan (multi-file changes, architectural decisions, security-sensitive logic, database schema changes), after completing a self-contained unit of work (module, endpoint, feature), when stuck or facing repeated failures (same test fails 2+ times, unexpected results), or when the agent wants independent validation of assumptions and design decisions. Triggers on any non-trivial implementation task where independent critique would catch blind spots before they become costly mistakes.

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。

Guideline for designing, implementing, and verifying secure APIs following OWASP API Security Top 10 (2023) best practices. Use when the user wants to: (1) review API code or design for security vulnerabilities, (2) design a secure REST, GraphQL, or gRPC API architecture, (3) implement API authentication and authorization (OAuth2, JWT, API keys, mTLS), (4) configure rate limiting, input validation, or CORS, (5) audit API endpoints for BOLA, BFLA, or mass assignment vulnerabilities, (6) create API security checklists or verification plans, (7) fix API security bugs or harden existing APIs, (8) set up API security testing (OWASP ZAP, Schemathesis, Burp Suite), or (9) handle any API security concern including SSRF prevention, resource consumption limits, business flow protection, API inventory management, and secure third-party API consumption.

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。

Generates Spring Boot 3.x configurations, creates REST controllers, implements Spring Security 6 authentication flows, sets up Spring Data JPA repositories, and configures reactive WebFlux endpoints. Use when building Spring Boot 3.x applications, microservices, or reactive Java applications; invoke for Spring Data JPA, Spring Security 6, WebFlux, Spring Cloud integration, Java REST API design, or Microservices Java architecture.

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。

Prepares Go applications to work effectively with AI coding agents. Use when setting up a new Go project or retrofitting an existing one to ensure: structured JSON logging (slog/Zap/ZeroLog/Logrus), machine-readable command output, thorough golangci-lint configuration, non-interactive CLI design with --yes flags, structured error handling with meaningful exit codes, proper context.Context propagation, graceful shutdown, health check endpoints, and a standardized Makefile. Triggers when a user asks to make their Go app "agent-ready," "AI-friendly," wants to improve agent tooling/observability in a Go project, or needs to audit an existing Go project against agent-readiness best practices.

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 serverless inference on Fireworks AI — open models, fine-tuning, and function-calling optimized endpoints.

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。

FastAPI guidance — typed endpoints, Pydantic validation, dependency injection, async patterns, and production deployment.

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。

Deploy and operate endpoint detection and response — policy tuning, behavioral detections, threat hunting, and response actions.

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 data loss prevention — data discovery, classification, policy enforcement across endpoint, network, and cloud.

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。

Align language models with human preferences using Direct Preference Optimization — no reward model, no RL loop.

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。