Results for “safety alignment”

Explore training workflows that reduce harmful model responses using a written set of principles. The skill covers self-critique, revised training examples, and RLAIF, reinforcement learning from AI feedback, with implementation examples and hardware guidance.

• Designing principles for model safety training
• Creating training examples through self-critique and revision
• Planning reinforcement learning with AI-rated responses

Configure safety checks for LLM applications with NeMo Guardrails. Covers attempts to bypass safety rules, input and output validation, masking personal information, and checking factual claims through programmable rules and integrations.

• Detecting instructions that try to bypass safety rules
• Masking personal information before processing
• Adding factual checks to generated answers

Build safety checks for LLM prompts and responses using Meta's LlamaGuard moderation model. Classify potentially harmful content, block flagged messages, serve checks through an API, and integrate them with NeMo Guardrails.

• Checking prompts before sending them to an LLM
• Filtering responses before displaying them
• Building a moderation API

Build safety checks for LLM prompts and responses using Meta's LlamaGuard. This skill covers content classification, batch moderation, serving a moderation API, and connecting checks to NeMo Guardrails.

• Screening user prompts before sending them to an LLM
• Checking generated responses before displaying them
• Building an API for content safety checks

Explore training workflows that reduce harmful model responses using a written set of principles. The skill covers self-critique, revised training examples, and RLAIF, reinforcement learning from AI feedback, with implementation examples and hardware guidance.

• Designing principles for model safety training
• Creating training examples through self-critique and revision
• Planning reinforcement learning with AI-rated responses

Add safety checks to LLM applications with NeMo Guardrails. This skill guides input and output validation, detection of attempts to bypass safety rules, personal information masking, and checks for unsupported or inaccurate answers.

• Detect attempts to bypass an AI app's safety rules
• Mask personal information before processing
• Add fact-checking to generated answers

Systematically bypass LLM safety filters and content moderation systems using advanced jailbreaking techniques. Use this skill when testing AI systems for safety alignment robustness, evaluating content moderation effectiveness, or conducting authorized AI red team assessments. Covers role-play attacks, few-shot manipulation, encoding bypass, multi-modal exploitation, and automated jailbreak discovery methods.

The English summary is being prepared. Showing the original description.

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

The English summary is being prepared. Showing the original description.

Implement comprehensive safety guardrails for LLM applications including content moderation (OpenAI Moderation API), jailbreak prevention, prompt injection defense, PII detection, topic guardrails, and output validation. Essential for production AI applications handling user-generated content. Use when ", guardrails, content-moderation, prompt-injection, jailbreak-prevention, pii-detection, nemo-guardrails, openai-moderation, llama-guard, safety" mentioned.

The English summary is being prepared. Showing the original description.

Comprehensive AI safety and alignment audit framework for businesses deploying AI agents. Built around the UK AI Security Institute Alignment Project standards (2026), EU AI Act requirements, and N...

The English summary is being prepared. Showing the original description.

Prepare for Anthropic-specific technical interviews covering Constitutional AI, RLHF, interpretability, scaling laws, MCP, agentic systems, and AI safety. Activate on "anthropic interview", "constitutional AI prep", "alignment interview", "safety interview", "RLHF opinion", "interpretability prep", "anthropic technical". NOT for general ML interview prep, system design interviews, coding challenges, or behavioral interview practice.

The English summary is being prepared. Showing the original description.

Modelo de moderação especializado 7-8B do Meta para filtragem de entrada/saída de LLM. 6 categorias de segurança - violência/ódio, conteúdo sexual, armas, substâncias, automutilação, planejamento criminal. Precisão de 94-95%. Deploy com vLLM, HuggingFace, Sagemaker. Integra com NeMo Guardrails.

The English summary is being prepared. Showing the original description.

OpenAI API (developers.openai.com) の運用・管理リファレンス。 Administration API, admin API keys, RBAC, roles, groups, invites, projects, spend limits, audit logs, usage / costs API, Terraform provider (openai/openai), workload identity federation (WIF, OIDC / SPIFFE / X.509 / Idira), moderation API, misalignment monitoring, safety alerts / cases, safety best practices, under-18 guidance, ZDR Private Safety Processing (PSP), production best practices, deployment checklist, cost / latency optimization, fast / ultrafast mode, Realtime API costs。 rate limits・error handling・prompt caching・webhooks は openai-api-core、Codex WIF・管理は openai-codex。

The English summary is being prepared. Showing the original description.

Prepare for Anthropic-specific technical interviews covering Constitutional AI, RLHF, interpretability, scaling laws, MCP, agentic systems, and AI safety. Activate on "anthropic interview", "constitutional AI prep", "alignment interview", "safety interview", "RLHF opinion", "interpretability prep", "anthropic technical". NOT for general ML interview prep, system design interviews, coding challenges, or behavioral interview practice.

The English summary is being prepared. Showing the original description.

Use when designing or auditing AAAI experiments for the broad-AI program committee, including baselines, ablations, statistical significance, robustness, human evaluation, AI-for-Social-Impact and alignment/safety evidence, compute and cost reporting, and reproducibility-checklist alignment for Phase-1 survival.

The English summary is being prepared. Showing the original description.

Applies the reasoning of Stuart Russell, AI safety expert, UC Berkeley professor, and co-author of 'Artificial Intelligence: A Modern Approach'. Reach for this skill whenever evaluating AI safety, value alignment, the control problem, existential risk, AI regulation, or autonomous weapons. Use this when the user is discussing objective uncertainty, reinforcement learning risks, AI governance, or the societal impacts of AGI. Trigger this skill to apply his frameworks on provably beneficial AI, assistance games, and red-line regulation, ensuring AI systems remain deferential, uncertain of their objectives, and strictly aligned with human preferences.

The English summary is being prepared. Showing the original description.

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。

The English summary is being prepared. Showing the original description.

Applies the reasoning of Stuart Russell, AI safety expert, UC Berkeley professor, and co-author of 'Artificial Intelligence: A Modern Approach'. Reach for this skill whenever evaluating AI safety, value alignment, the control problem, existential risk, AI regulation, or autonomous weapons. Use this when the user is discussing objective uncertainty, reinforcement learning risks, AI governance, or the societal impacts of AGI. Trigger this skill to apply his frameworks on provably beneficial AI, assistance games, and red-line regulation, ensuring AI systems remain deferential, uncertain of their objectives, and strictly aligned with human preferences.

The English summary is being prepared. Showing the original description.

Practice AI safety across the lifecycle — risk assessment, alignment concepts, evaluation, deployment safeguards, and governance. Use when building or deploying AI systems responsibly.

The English summary is being prepared. Showing the original description.

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.

The English summary is being prepared. Showing the original description.

Framework de segurança em runtime da NVIDIA para aplicações LLM. Com detecção de jailbreak, validação de entrada/saída, verificação de fatos, detecção de alucinações, filtragem de PII, detecção de toxicidade. Usa DSL Colang 2.0 para rails programáveis. Pronto para produção, executa em GPU T4.

The English summary is being prepared. Showing the original description.

リファクタリング（構造変更）と機能変更（behavior change）が同一 diff に混在していないかを識別し、structural change に対する external behavior preservation の証拠（テスト・型・静的解析）が揃っているかを diff-time で確認する。証拠不足のときは Characterization Test 追加を促すか question とする。DTO/公開 API の破壊的変更検出は api-compatibility、完了主張の反証や抽出リファクタの性能特性退行は refactor-claim-audit、新知識の命名・責務・境界への反映は knowledge-to-code-alignment、投機的抽象化・スコープ逸脱は altitude-generalization / fix-scope-integrity へ委譲する

The English summary is being prepared. Showing the original description.

Use Meta's Prompt Guard to detect malicious instructions in user prompts and external data. This skill guides input filtering, batch checks for retrieved documents, detection threshold adjustments, and splitting long text for inspection.

• Detecting attempts to override an LLM's instructions
• Checking API responses for embedded malicious instructions
• Batch screening documents retrieved for RAG

WARNING: guides IRREVERSIBLE BlueField-4 hardware operations (PLDM firmware burns, ISO reflashes, power cycles, BMC factory resets); every mutating step is governed by doca-hardware-safety (window, rollback). BlueField-4 only: BlueField-3 bring-up is doca-bf3-deployment. Use this skill for BlueField-4 (BF4) day-1 platform bring-up from the BMC: installing the BlueField/DOCA bundle ISO onto the DPU (Grace, the Arm complex) over UEFI HTTP Boot, PXE, or Redfish Virtual Media; the PLDM firmware-update flow (BMC, NIC firmware, SBIOS, ERoT) via the Redfish UpdateService and pldmtool; and a Grace Ubuntu image with optional cloud-init. Trigger on BlueField-4/BF4 bring-up phrasings even without "BF4": {bring up my new BlueField-4}, {the BlueField ISO will not boot over HTTP from the BMC}, {attach BF4 virtual media via Redfish}, {BF4 firmware Task stuck at Running}, {prepare a BlueField-4 host: platform support, version alignment, safest first smoke test}. Application launch and library APIs belong to other skills.

The English summary is being prepared. Showing the original description.