「ai safety」の検索結果

Implements input and output validation guardrails for LLM-powered applications to prevent prompt injection, data leakage, toxic content generation, and hallucinated outputs. Builds a security validation pipeline using NVIDIA NeMo Guardrails Colang definitions, custom Python validators for PII Tespit and content policy enforcement, and the Guardrails AI framework for structured output validation. The guardrails system intercepts both user inputs (blocking injection attempts, stripping PII, ...

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。

Safety-case design for always-on agents with episodic memory. Covers data hygiene, privacy and security risk surfaces, cost controls, scope, user dependency concerns, retention, incident containment, and responsible deployment evidence. It identifies hazards and asks for named controls, tests, owners, and residual uncertainty; it does not provide legal, clinical, youth-safety, or jurisdictional determinations. Activate on: "agent safety", "always-on agent privacy", "agent cost control", "persistent agent risks", "AI companion safety", "agent data hygiene", "runaway agent costs", "parasocial AI risk", "/always-on-agent-safety". NOT for: architecture design (use always-on-agent-architecture), input design (use always-on-agent-inputs), application brainstorming (use always-on-agent-applications), healthcare compliance specifically (use hipaa-compliance).

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。

Deploys Llama Guard 3 safety classification, NeMo Guardrails programmable dialogue rails, and LLM Guard input/output scanner pipelines as complementary runtime defenses that inspect and constrain LLM prompts and responses. Use when adding a production runtime safety layer to an LLM, RAG, or agent application to block jailbreaks, prompt injection (OWASP LLM01), toxic content, or sensitive-data leakage before it reaches or leaves the model.

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。

Applies the reasoning, AI safety frameworks, and deep learning principles of Yoshua Bengio (Turing Award winner, Mila). Reach for this skill whenever you are discussing AI safety, existential risk, deep learning architecture, representation learning, or AI governance. Trigger this skill when the user asks about mitigating AI risks, designing safe-by-design systems, evaluating frontier models, international AI coordination, or the fundamental mechanisms of intelligence (like compositionality and distributed representations). Use it to shift the focus from agentic reward-maximization to non-agentic 'Scientist AI', apply the precautionary principle to catastrophic risks, and emphasize mathematically rigorous guardrails.

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。

Azure AI Content Safety SDK for Java workflow skill. Use this skill when the user needs Build content moderation applications using the Azure AI Content Safety SDK for Java and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.

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。

Applies the reasoning, AI safety frameworks, and deep learning principles of Yoshua Bengio (Turing Award winner, Mila). Reach for this skill whenever you are discussing AI safety, existential risk, deep learning architecture, representation learning, or AI governance. Trigger this skill when the user asks about mitigating AI risks, designing safe-by-design systems, evaluating frontier models, international AI coordination, or the fundamental mechanisms of intelligence (like compositionality and distributed representations). Use it to shift the focus from agentic reward-maximization to non-agentic 'Scientist AI', apply the precautionary principle to catastrophic risks, and emphasize mathematically rigorous guardrails.

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。

Test AI agent systems for tool abuse, unauthorized actions, privilege escalation through tool chaining, and safety bypass via agentic workflows. Use this skill when assessing autonomous AI agents that use tool-calling (function calling, plugins, actions) to interact with external systems. Covers multi-step attack chains, implicit trust exploitation, and capability boundary testing for AI 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。

Método da Anthropic para treinar IA inofensiva através de auto-aprimoramento. Abordagem em duas fases - aprendizado supervisionado com auto-crítica/revisão, depois RLAIF (RL from AI Feedback). Use para alinhamento de segurança, redução de outputs prejudiciais sem rótulos humanos. Alimenta o sistema de segurança do Claude.

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。

Knowledge base from FAA-HDBK-006D (2020), the FAA's System Reliability, Maintainability, and Availability (RMA) Handbook for National Airspace System (NAS) hardware and software. Use for RMA figures of merit (MTBF, MTBO, MTTR, the availability variants), the bathtub curve and probability distributions, the four-stage RMA lifecycle mapped to the FAA Acquisition Management System (AMS), the six-task RMA technical-management/acquisition process, service-thread criticality and the NAS-RD severity-to-target table, the RMA analysis toolbox (RBD, FMEA/FMECA, FTA, Ishikawa, Monte Carlo, Bayesian, FRACAS, ALT, reliability-growth/recovery tests), software reliability (early prediction + reliability growth), and tailorable example RMA requirements for a Program Requirements Document. Scope limits: it is FAA/NAS-specific guidance (not a requirement, by its own statement), centred on hardware-plus-software availability — it brackets out human/service/facility/communications availability, names but does not reproduce the math derivations or external standards (MIL-STDs, IEEE 1633, RTCA DO-178C/DO-278A, SAE JA1011), is thin on detailed worked numerical examples and on safety-case method, and predates any current revision beyond 006D.

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 Safety Specialist IA — Expert en sécurité IA (alignment, guardrails, filtrage de contenu, test de biais, red teaming)

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。

NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.

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。

Decide which AI features to build, delay, shelve or kill, and whether each one gets better as models improve. Produces an AI Bet Memo (feature-by-feature verdicts, retest triggers, scaffolding ledger, moat and safety check) from a roadmap or idea list. Use for 'AI strategy', 'AI roadmap', 'what should we build with AI', 'will the next model make this obsolete', 'build vs wait for the model', 'will OpenAI/Anthropic build this', 'agent or chatbot', 'is this feature worth shipping'. Draws on 118 Lenny's Podcast guests incl. Nick Turley, Kevin Weil and Dianne Penn.

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。

当需要为 AI 系统设计多层安全防线、内容过滤策略和伦理边界时调用此 skill。典型场景包括：设计拒绝策略与升级机制、防御 prompt 注入攻击、实现领域特定安全规则（教育、医疗、金融等）、定义 AI 的价值观锚点。 不适用于：定义工具操作权限（应使用 tool-specification）、定义 AI 身份（persona-design）、调整输出风格（personality-system）。 关键 trigger 信号：AI 涉及敏感话题、需要设计"拒绝回答"策略、存在 prompt 注入风险、特定行业合规要求、需要多层防御而非单一规则。

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。

Azure AI Content Safety SDK for Python workflow skill. Use this skill when the user needs Azure AI Content Safety SDK for Python. Use for detecting harmful content in text and images with multi-severity classification and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.

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。

This skill should be used when the user asks to "buy token", "sell token", "swap token", "trade crypto", "check trade status", "query transaction", "scan tokens", "feed", "monitor chain", "query token", "token details", "check token safety", "list wallets", "show wallets", "my wallets", "AI scan", "AI扫链", "auto scan", "smart scan", "tweet scan", "推文扫链", "twitter scan", "onboarding", "get started", "check IP", "get IP", "IP whitelist", "查IP", "IP白名单", "launch token", "create token", "发币", "创建代币", "auto sell", "stop loss", "take profit", "trailing stop", "止盈止损", "移动止损", "holders", "top holders", "kol holders", "insider", "持仓", "持仓列表", or mentions trading on Solana/ETH/BSC/Base chains via XXYY. Enables on-chain token trading and data queries through the XXYY Open API.

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。

Evaluate current trail conditions including weather, snow line, river crossings, exposure, and trail maintenance status for safety decision-making. Produces a GREEN/YELLOW/RED safety rating with actionable go/no-go recommendations. Use the day before or morning of a planned hike, during tour planning to assess seasonal viability, after unexpected weather changes on a multi-day tour, when reports suggest trail damage or closures, or before committing to an alpine or exposed route.

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 reliable, safe, and trustworthy agent systems that fail gracefully and operate within bounds. Use when building guardrails, handling edge cases, preventing harmful outputs, monitoring for failures, implementing safety constraints, or designing error recovery. Covers failure modes, constraint systems, oversight mechanisms, and safety validation.

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。

Test Large Language Models for direct prompt injection vulnerabilities where user input overrides system instructions, extracts system prompts, bypasses safety filters, or causes unauthorized actions. Use this skill when assessing chatbots, AI assistants, LLM-powered tools, or any application that processes natural language input through an LLM. Covers role-playing attacks, instruction hierarchy exploitation, multi-turn manipulation, and context window abuse for comprehensive AI security testing.

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。

Configures best-practice alerting policies for AI agents using OpenTelemetry (OTel) metrics, generating output as Terraform (.tf) configuration files. Use when analyzing, writing, or deploying alerting policies to monitor agent latency, error rates, token usage, and quality metrics. Don't use for standard infrastructure monitoring unrelated to AI agents, or when the agent is not instrumented with OpenTelemetry (for Reliability, Cost, Safety, Security alerts). NOTE: Reliability, Cost, Safety, and Security alerts use generic OTel metrics and work across runtimes (such as Cloud Run, Vertex AI). Quality alerts rely on Vertex AI Online Monitors and are strictly bound to Vertex AI deployments.

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。

Configure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI model governance, MCP rate limiting, jailbreak detection, add Azure OpenAI backend, add AI Foundry model, test AI gateway, LLM policies, co...

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エージェントの危険なコマンドを実行前に確認し、ファイル編集を指定ディレクトリ内に制限します。単独または併用のモードで、本番作業や自律実行時の操作を管理します。

• 本番作業で危険なコマンドを確認
• 特定ディレクトリだけ編集したいとき
• 自律実行時に保護を併用したいとき

Patterns and techniques for adding governance, safety, and trust controls to AI agent systems. Use this skill when: - Building AI agents that call external tools (APIs, databases, file systems) - Implementing policy-based access controls for agent tool usage - Adding semantic intent classification to detect dangerous prompts - Creating trust scoring systems for multi-agent workflows - Building audit trails for agent actions and decisions - Enforcing rate limits, content filters, or tool restrictions on agents - Working with any agent framework (PydanticAI, CrewAI, OpenAI Agents, LangChain, AutoGen)

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。

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

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 AI agent code safely in isolated sandboxes with resource limits, audit trails, and kill switches. Use when someone asks to "sandbox my agent", "run agent code safely", "add guardrails to AI agent", "isolate agent execution", "audit agent actions", "prevent agent from deleting files", "restrict agent permissions", or "add safety controls to AI coding agent". Covers Docker isolation, filesystem restrictions, network policies, resource locking, and comprehensive audit logging.

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。