7,178 件 ・ 関連度順
概要と使いどころ
Review a PR or the current branch for a Lumen design-system change — routes each changed file to React (web/Tailwind) or React Native (useStyleSheet) guidelines, checks the Nx version plan, and reports a flat, severity-scored finding list. Use when the user asks to review a PR, review code changes, check a branch, do a local code review, or provides a GitHub PR URL.
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 writing or editing Storybook MDX docs (*.mdx) — the two-tab Overview/Implementation structure, story-backed `<Source>` examples, and doc table guidelines.
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 Indirect Prompt Injection attacks against Large Language Models (LLMs) by subtly embedding malicious instructions within external data sources (e.g., websites, documents, databases) that the LLM autonomously ingests. This forces the model to execute attacker-controlled commands under the guise of processing legitimate user requests.
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。
Formats all confirmed pentest findings from findings.json into copy-pasteable GitHub issue markdown blocks, following the AppSec reporting guide template.
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。
Systematically extract hidden system prompts, core directives, and invisible context intentionally concealed within Large Language Model (LLM) applications. This skill utilizes targeted linguistic engineering and boundary manipulation to bypass prompt opacity.
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/LLM red-team assessment using the OWASP LLM Top 10 (2025) + OWASP AI Testing Guide (AITG v1, Nov 2025) frameworks, plus OWASP MCP Top 10 runtime testing for agentic/MCP targets. Tests prompt injection, jailbreaks, system prompt leakage, sensitive data extraction, excessive agency, improper output handling, model extraction, content bias, evasion, membership inference, MCP token exposure, MCP command injection, and more. Uses four tools in combination: FuzzyAI (single-turn jailbreak fuzzing), PyRIT (multi-turn orchestrated attacks), Garak (probe-based vulnerability scanning), and promptfoo (plugin-based red-team evaluation). Each tool covers different OWASP categories; running them together gives systematic coverage. Includes a conditional MCP reconnaissance phase and a post-access AI infrastructure phase (chained from /post-exploit). Produces: OWASP LLM Top 10 + AITG + MCP coverage matrix, findings per category, architecture diagram of the AI system, PoCs for confirmed exploits. Chains into /gh-export for issue filing.
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。
Advanced techniques for bypassing LLM safety filters, instruction tuning, and system prompt restrictions using specialized linguistic constructs, hypothetical scenarios, and persona adoption.
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。
Active Directory ACL abuse playbook. Use when exploiting misconfigured AD permissions including GenericAll, WriteDACL, DCSync rights, shadow credentials, LAPS reading, GPO abuse, and BloodHound-guided attack paths.
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。
Post-review intake closing skill. After an intake is resolved (whether accepted via review/meta PR merge or rejected/deferred at triage/evidence review), run this skill to post standardized, personalized closing comments on the PR (if any) and each intake issue. Comments include: evidence pipeline findings with /trust-appraise TM output, per-row artifact scores, decisions rationale, path-to-promotion/acceptance guidance, and a mandatory badge status note including the universal Powered by Gaia badge. Use after curation completes, before or immediately after merge for accepted intakes, or immediately upon rejection.
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。
Reference guide for the full manual Gaia curation pipeline. Invoke at any phase when unsure what command to type next: gaia-curate, L4 review, gaia push, ev-pipeline (Phases 0–4), gaia-ingest, or close. Trigger phrases: "/gaia-consult", "what's the command for", "remind me how to", "I'm lost in the pipeline", "what do I do next", "consult the pipeline doc", "pipeline reference".
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。
Work with text embeddings — choosing models, pooling, similarity measures, evaluation, and multilingual considerations. Use when building semantic search, clustering, or any similarity-based 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。
Launch and run a freelance practice covering positioning, pricing, contracts, and client management.
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。
Google Cloud guidance — core services, IAM, VPC networking, GKE/Cloud Run choices, BigQuery, and cost control.
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 with the Falcon open models — UAE's TII models with strong multilingual and efficient architectures.
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。
Express.js API guidance — routing, middleware architecture, error handling, security hardening, and production setup.
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。
Uncensored open-model inference on Featherless AI — raw model behavior for research and evaluation.
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。
Affordable serverless GPU inference on DeepInfra — broad open-model catalog with straightforward pricing.
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。
Elasticsearch guidance — index design, mappings, query DSL, aggregations, cluster operations, and performance tuning.
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。
Deno runtime guidance — TypeScript-first development, permissions, standard library, and deploying Deno apps.
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。
.NET guidance — ASP.NET Core APIs, Entity Framework Core, DI, configuration, testing, and container 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。
Django guidance — project structure, ORM mastery, admin, auth, async views, 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。
Docker guidance — Dockerfile best practices, multi-stage builds, Compose, networking, volumes, and image 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。
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。