「token」の検索結果

Use when adding support for a new model architecture to imp, porting a model family, or debugging a model that loads but produces wrong output - "add support for <model>", "new arch", loader detection, chat template, tokenizer parity, RoPE variant, "outputs garbage", "prompt-blind", "digits scrambled", "NaN logits", "describes a different picture", "does it fit in VRAM". Do NOT use for kernel performance (sm120-cuda-expert) or quant-format questions (quant-formats).

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 imp outputs differ between two paths that should agree - chunk sizes, prefill vs decode, prefix-cache resend, two engines (imp vs llama.cpp vs HF), two builds or kernels - "output changes with chunk size", "first token differs", "logprobs moved", "which layer diverges", "who is right, imp or llama.cpp", "near-tie". Do NOT use for repetition loops or garbage output (check-degeneration), a new arch that loads wrong (add-model-arch), or perf (benchmark-cuda).

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。

All-in-one fullstack dev engine. /aura: 46 modes (build/fix/clean/deploy/review/spec/lore/ax/experiment/payment/debug/qa/orchestrate/escalate+), 6-layer security with 32 hooks, tiered models (ZERO/ECO/PRO/MAX), 8 languages, 16 specialized agents, SPEC/EARS/TRUST5/XLOOP/RALF/Autopus absorbed. ~55% token savings.

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 templates by replacing {{TOKEN}} placeholders with actual values, supporting all three templates (specification, plan, tasks) with schema validation and security sanitization

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。

Configurable regex-based alert system for detecting patterns in tool calls and session activity. Supports error triggers, content regex matching, token threshold triggers, and pattern detection with configurable 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。

Decision matrix for selecting multi-agent topologies (Supervisor, Swarm, Hierarchical, Conductor) with token economics, failure modes, and escalation 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。

Fintech engineering expertise — payment processing (Stripe, Plaid), PCI DSS compliance, financial data modeling (double-entry bookkeeping), fraud detection patterns, bank-grade security (encryption, tokenization), open banking APIs, cryptocurrency/blockchain integration, regulatory compliance (KYC/AML), and idempotent financial transaction design. Use for payment systems, banking apps, trading platforms, and fintech infrastructure.

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。

Workflow for repository reconnaissance and operations using GitHub CLI (gh). Optimizes token usage by using structured API queries instead of blind file fetching.

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 Git operations wrapper. Optimizes token usage by guiding complex git workflows into efficient CLI commands.

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。

The Agent Tool Contract — 5 principles for designing tools agents call reliably: predictable signature, rich errors, token-efficient output, idempotency, graceful degradation. Includes anti-pattern table with 8 common 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。

OAuth 2.1 compliant authentication flows (MANDATORY Q2 2026). PKCE required for ALL clients, Implicit Flow removed, modern token 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。

Hunt full Account Takeover (ATO) by chaining lower-severity findings — IDOR-to-email-disclosure, password-reset-token-theft, OAuth-redirect-uri abuse, response-manipulation, cookie scope, CRLF in Set-Cookie, race-condition on signup. Use when the user wants to escalate small findings to ATO or thinks an ATO path exists.

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。

PDF / 图片 / 网页高质量解析为 Markdown。三种部署形态全部免费：本地 CLI（开源）、 云 API（注册免费 token + 配额，超额才付费）、网页 flash（零注册，单文件 ≤10MB / ≤20页）。 在中文学术 PDF / 表格 / 公式上识别质量显著高于 pdfplumber。

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 wants to capture a web page / article / docs into their second brain vault as clean Markdown — stripping ads, navigation, and sidebars to save tokens. Trigger on "clip this URL", "save this article to my vault", "ingest this page into raw/", or when gathering source material for the knowledge base.

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。

Convert Figma designs into production-ready code using MCP server tools. Use this skill when users provide Figma URLs, request design-to-code conversion, ask to implement Figma mockups, or need to extract design tokens and system values from Figma files. Works with frames, components, and entire design files to generate HTML, CSS, React, or other 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。

Slack workspace integration: 9 on-demand scripts (post, read, delete, search, stream, react, upload, channels, users) + dual-token support + RTS API search + chat streaming + AI Block Kit + Session Bridge + Claudicle unified launcher (Claude Agent SDK, soul engine, three-tier memory).

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 and prototype in Paper Design (app.paper.design) — read/write designs via MCP, convert to React/Tailwind code, import HTML/CSS, sync design tokens. Also covers Paper Shaders (@paper-design/shaders-react) — 30+ shader React components (mesh gradient, fluted glass, liquid metal, god rays, halftone, grain gradient, pulsing border). Pairs with minoan-frontend-design for creative direction. Triggers on Paper Design, paper.design, design-to-code, artboard, Paper Shaders, Paper MCP, prototype.

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 frontend code drafts using Gemini 3.1 Pro's one-shot UI generation. Use when building React components, HTML/CSS pages, SVG animations, or converting screenshots to code. Produces fast, cheap drafts (7.5x cheaper than Opus) that Claude reviews and refines. Supports design system context loading (1M token window), screenshot-to-code pipeline, and physics-based SVG animation. Trigger: "generate a UI", "screenshot to code", "draft a component", "SVG animation", "convert this design", "gemini forge", "forge a UI".

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。

Evaluates whether work-in-design is ready to hand off from a chat design conversation to autonomous execution (Claude Code, n8n, agent runtime). Runs a 7-condition gate: spec stability, verification surface, invariants, pump-primer, work decomposition, rollback cost, token budget headroom. Three invocation modes — deliberate (caller provides work_context, returns JSON verdict), proactive sensing (Claude detects handoff signals in design conversation and offers the check), conversational walkthrough (Claude walks the 7 conditions as discussion topics). Returns structured JSON in all modes. Use when designing a Claude Code deployment, scoping an autonomous run, or deciding next move. Trigger on: "ready to hand off," "ready for execution," "is this CC-ready," "handoff check," "walk me through readiness," "are we ready to ship this," "should I just hand this off," or proactive detection. Do NOT use for evaluating completed work, runtime selection, or prompt quality. Profile-agnostic.

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 Claude Project context budget under the automatic-RAG-by-window model: knowledge files vs. threshold-exempt overhead (Skills, MCPs, CI, Memory). Two modes: Quick Diagnostic and Full Budget Audit. Use when user says "check my context budget," "how much context am I using," "is my project too big," "optimize my token usage," "tier my files," "optimize for RAG," "improve retrieval quality," "should I keep compressing," "am I over-compressing," "should I accept RAG mode." Also trigger on context pressure symptoms: "Claude forgets my instructions," "responses getting generic," "content not found in my knowledge files." Also use when a project audit scores Knowledge Architecture ≤ 3. Do NOT use for content placement decisions (use rootnode-memory-optimization if available), full project audits (use rootnode-project-audit if available), or behavioral tuning (use rootnode-behavioral-tuning if available). Run on Opus 5 or Sonnet 5 at `high` effort (both defaults); depth reduces on legacy models.

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 working with iOS/macOS Keychain Services (SecItem queries, kSecClass, OSStatus errors), biometric authentication (LAContext, Face ID, Touch ID), CryptoKit (AES-GCM, ChaChaPoly, ECDSA, ECDH, HPKE, ML-KEM), Secure Enclave, secure credential storage (OAuth tokens, API keys), certificate pinning (SecTrust, SPKI), keychain sharing across apps/extensions, migrating secrets from UserDefaults or plists, or OWASP MASVS/MASTG mobile compliance on Apple platforms.

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。

审视、设计、从零启动或改进 HarmonyOS、OpenHarmony 与 ArkUI 产品界面。用于检查 ArkTS UI、建立新项目设计基线、区分永恒原则与当前平台视觉语言、规划跨手机/平板/折叠屏/PC 的导航和布局、多输入控件状态、animateTo、transition、geometryTransition、springMotion、手势跟手与可打断动效、异步状态真实性、视觉分层参数、无障碍或动画性能，或建立鸿蒙 Motion Token 时。不要用于签名、构建、网络、存储、权限、数据库等纯工程问题，也不要用于通用视频剪辑或其他移动平台的设计规范评审。

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 skill when tool outputs exceed 2000 tokens, tasks span multiple conversation turns, sub-agents need to share state, context window is bloating, plans need to persist across summarization, terminal/log output needs selective querying, or when user mentions "offload context", "dynamic context discovery", "filesystem memory", "scratch pad", "reduce context bloat", or "just-in-time context loading".

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 or review a Claude Design brief and convert its output into a deterministic SwiftUI implementation handoff with native iOS layout, tokens, states, transitions, and accessibility rules. Use for Claude Design mobile prototypes, three-screen iOS concepts, Design-to-Claude-Code handoffs, or when a visual prototype must be buildable without interpretation.

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。