AI & ML

llm-coding Bundle

Install Semantic Contracts (and, as a sub-mode, individual semantic anchors) into a coding agent's persistent context for Claude Code, Codex, Gemini CLI, Cursor, GitHub Copilot, and similar agents. Use when setting up project-level or user-level onboarding, generating or updating AGENTS.md / CLAUDE.md / GEMINI.md / Copilot instruction files, or choosing which contracts a project should run. A contract is the part the LLM cannot guess — a named composition of anchors plus the project's own rendered template — so installing contracts is the high-value path; installing a bare anchor list is the lightweight fallback.

diverta Bundle

Kurocoサイトのセキュリティ設定を Admin MCP 経由で読み取り、チェックリストに照らしてリスクを診断・報告する読み取り専用スキル。設定の変更は一切行わない。APIセキュリティ方式・CORS・IPアドレス制限・ログイン/パスワードポリシー・2要素認証（ワンタイムパスワード）・権限グループとスーパーユーザー一覧・静的アクセストークン/シークレットの棚卸し・公開されているAPI・監査ログを対象とする。「セキュリティチェック」「セキュリティ監査・診断・レビュー」「設定の棚卸し」「security audit」「CORS/IP制限/権限を確認」「セキュリティチェックシート」「脆弱性が心配」など、Kurocoの設定が安全か点検・報告する依頼で使用。

diverta Bundle

Kuroco の API パフォーマンスとコストを Admin MCP の読み取りツール（usage-get / api_analytics-list / api_log-list / api_uri-list）で調査し、キャッシュ設定を中心とした改善提案をまとめる。利用料が急に増えた、APIリクエスト課金が高い、キャッシュヒット率が低い、MISS/PASS が多い、特定エンドポイントのリクエストが急増した、レスポンスが遅い・実行時間が長い、CDN転送量が増えた、クローラーの負荷を調べたい、といったコスト・パフォーマンスの原因調査とレポート作成に使用する。

diverta Bundle

Methodology for iteratively improving agent-facing instructions (skills / slash commands / CLAUDE.md / code-gen prompts) via bias-free executor + two-sided evaluation (self-report + instruction-side metrics). Meta-skill, invoke ONLY when the user explicitly asks for an "empirical" eval of a prompt or skill, or for the Iter-0 description / body consistency check. Do NOT auto-invoke after every skill edit; this loop is operator-triggered by name.

lukasstrickler Bundle

Search the web, library documentation, and GitHub repositories using Tavily, Context7, GitHub Grep, Exa fallback, and Z.AI Web Search Prime MCP fallback. Use when: (1) Looking up documentation for libraries or frameworks, (2) Searching for code examples or tutorials, (3) Finding API references or specifications, (4) Researching best practices or solutions, (5) Looking up error messages or troubleshooting guides, (6) Finding library installation instructions, (7) Searching for real-world code patterns in GitHub repositories, or (8) When you need current web information or documentation. Triggers: search, look up, find documentation, search web, lookup, find examples, search for, how to, tutorial, API reference, documentation for, error message, troubleshoot, best practices, find code examples, GitHub search.

aomi-labs Bundle

Build natural-language crypto agents, web3 assistants, and trading bots that read and write EVM chain state. Aomi turns prompts ("swap 1 ETH for USDC", "open a 3x GMX long", "bet $100 on Polymarket") into wallet-signed transactions on Ethereum, Base, Arbitrum, Optimism, Polygon, Linea — non-custodial, fork-simulated. Use when the user wants a crypto/DeFi agent, AI trading/wallet assistant, or on-chain execution against Uniswap, Aave, Lido, GMX, Hyperliquid, Polymarket, Binance, OKX, or 40+ other protocols. Trigger with prompts about swaps, lending, bridging, staking, perps, prediction markets, or any DeFi/CEX action needing a wallet signature. Account-abstraction aware with EIP-7702/4337 in the backend lane; local CLI signing is EOA. MUST NOT fabricate or echo credentials; values reach the CLI only when the user explicitly supplied them.

aomi-labs Bundle

Aomi CLI — natural-language to executable on-chain transactions via an agent. Send a prompt like 'swap 1 ETH for USDC' and the agent stages calldata, simulates as a batch, and returns a queued tx for you to sign. Account-abstraction-first signing (EIP-7702 on mainnet, ERC-4337 on L2s) with EOA fallback. Covers swaps, lending, bridging, staking, perps, and CEX read across 25+ apps (Uniswap, Aave, CCTP, Across, Lido, GMX, Polymarket, Binance, etc.) on Ethereum, Polygon, Arbitrum, Base, Optimism, and more. Use when an agent should drive on-chain actions through a CLI rather than constructing calldata directly. Triggers: aomi, agent CLI, natural-language transaction, intent execution, simulate-then-sign.

aomi-labs Skill

GOAT (Great Onchain Agent Toolkit) — 200+ protocol integrations across 30+ chains. Tool creation, framework adapters (AI SDK/LangChain/Eliza), DeFi actions (swap/bridge/transfer), wallet management, and modular plugin architecture for building onchain AI agents.

aomi-labs Skill

HTTP 402 payment protocol for AI agent commerce — three-actor model (Client, Resource Server, Facilitator), ERC-3009 transferWithAuthorization, server middleware (@x402/express), client patterns in TypeScript and Python, facilitator integration, agent-to-agent payments, pricing strategies, and replay protection. Works on Base, Ethereum, Arbitrum, Optimism, Polygon, and Solana.

aomi-labs Skill

elizaOS multi-agent AI framework — character files, plugin system, platform connectors, trust scoring, RAG knowledge, and Solana wallet integration. Use when building autonomous AI agents with personality, multi-platform presence, or onchain capabilities.

aomi-labs Skill

Comprehensive guide for building AI agents that interact with Solana blockchain using SendAI's Solana Agent Kit. Covers 60+ actions, LangChain/Vercel AI integration, MCP server setup, and autonomous agent patterns.

aomi-labs Skill

Step-by-step guide for creating enriched CryptoSkills agent skills. Use when building new protocol skills, contributing to the directory, or understanding the enriched skill pattern. Covers SKILL.md structure, YAML frontmatter, examples, docs, resources, templates, marketplace registration, and validation. Triggers: "create a skill", "add a protocol", "contribute a skill", "new skill template".

nvidia-omniverse-ovstream Skill

CUDA buffer allocation, pitch alignment, BGRA channel order, and ctypes-bound cudart usage. Use when user asks about CUDA interop, pitched allocations, frame layout, or zero-copy GPU integration.

nvidia-omniverse-ovstream Skill

Submitting video frames to ovstream (raw CUDA BGRA8 or pre-encoded bitstreams). Use when user asks how to push frames, the video frame format, pitch alignment, or pre-encoded passthrough.

nvidia-omniverse-ovstream Skill

Writing a CUDASHM reader/consumer that attaches to an ovstream CUDASHM producer over CUDA IPC. Use when user asks about consuming GPU-resident frames, reading from a CUDA IPC ring, writing a simulation/post-processing kernel that ingests rendered output, or running a consumer in a separate container.

mozilla Bundle

Use this skill to audit tables for missing column descriptions and classify each missing column into the correct base schema promotion target (global.yaml, app_<product>.yaml, or <dataset_name>.yaml). Accepts a dataset name and an optional table filter — omit the filter to audit all tables in the dataset. Outputs a per-column recommended_target report for use in _missing_metadata.yaml. Composable with schema-enricher (Step 6).

mozilla Bundle

Use this skill when looking up, auditing, or managing column descriptions from global, application-specific, and dataset-specific column definition YAML files (bigquery_etl/schema/global.yaml, bigquery_etl/schema/app_<name>.yaml, and bigquery_etl/schema/<dataset>.yaml). Use it to find a description for a specific column, list all columns in a base schema, audit which columns in a table's schema.yaml are covered by base schemas, or identify columns missing descriptions. Works with schema-enricher skill.

openocta Bundle

Linux 与网络设备只读健康巡检；直接运行 scripts/patrol.sh，配置由平台侧信道注入，Agent 不读取环境变量

zhanghandong Bundle

CRITICAL: Use for managing agent memory based on CoALA cognitive architecture. Triggers on: remember this, 记住这个, save this info, learn from this, 学习这个, memory system, 记忆系统, knowledge base, what did we learn, 之前学到的, project context, 项目上下文, my preferences, 偏好设置, recall memory, 回忆一下, what do you know about, 你知道什么, forget this, 忘记这个, reflect on, 反思一下, summarize session, 总结会话

naveedtechlab Bundle

Enable AI-driven Todo management using natural language via MCP tools. Use when building conversational todo agents, implementing natural language task management, designing AI assistants for todo apps, or creating MCP-based tool integrations. Covers intent interpretation, safe tool invocation, and stateless agent design.

lucas-martino Bundle

Aligns technical development with the chosen commercial model (Closed Source, Open Source Proprietary, Hybrid or GPL).

lucas-martino Bundle

Guardian of **Model-Driven Design**. The primary objective is to capture the domain expert's mental model into the software.

rohunvora Skill

Research-driven UI/UX improvement using Mobbin. Use when user explicitly requests design pattern research ("use Mobbin", "research design patterns", "find UX patterns for X"). Requires Mobbin account and browser automation (claude-in-chrome MCP). Searches Mobbin for relevant UI patterns, extracts design principles, generates a spec document for approval, then implements. NOT for general UI work—only when user wants research-backed design.

michtio Bundle

Releasing Craft CMS plugins — tagging, Packagist propagation, GitHub releases, branch promotion, shared-library ordering, history rewrites. ALWAYS load when cutting, preparing, verifying, or debugging a plugin release: bumping a version, dating a changelog, creating or moving a git tag, or checking what Packagist serves. Covers the composer.json version key (bump-or-omit trade-off, same-commit rule, verifying the tag's own blob), Packagist verification via repo.packagist.org/p2, the 'Skipped tag' silent failure, the tag recreation risk model (unserved safe, served never), GitHub release objects drifting from tags, gh api PATCH wiping tag_name, blank bodies from empty --notes-file, the create-release.yml Store dispatch (422 already_exists, allowUpdates), shared-library releases dependency-first (additive minors breaking released consumers, module: track invalid), two-way origin comparison, filter-repo purges (already_ran re-runs, --refs/--partial, tree-hash verify, blob-level sweeps), and path repositories (ca

syou6162 Skill

「開発日誌更新」「開発日誌作って」の言及時に使用。esa-llm-scoped-guardで開発日誌を新規作成・更新します。

mvdmakesthings Bundle

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

thepm001 Bundle

Use this skill for AEP 2.8, dynAEP (main AEP event runtime), Base Node, Composer Lite, CCA / setup agent, CAW, UCB, Path A/B, dynAEP-TA, dynAEP-TA-P, Lattice Memory, Admit collect-all, scanners, evidence ledger and governed UI. Prefer SKILL.md (uppercase) as the full protocol skill. Public repo: https://github.com/thePM001/AEP-agent-element-protocol

thepm001 Skill

Use when creating a new AepCaw security policy, including agent sandboxes, CI pipelines, development environments, HTTP service gateways, or Postgres-family database access policies

shareai-lab-kode Skill

Evaluate Agent Skill design quality against official specifications and best practices. Use when reviewing, auditing, or improving SKILL.md files and skill packages. Provides multi-dimensional scoring and actionable improvement suggestions.

shareai-lab-kode Bundle

Transform an AI agent into a tasteful, disciplined development partner. Not just a code generator, but a collaborator with professional standards, transparent decision-making, and craftsmanship. Use for any development task: building features, fixing bugs, designing systems, refactoring. The human provides vision and decisions. The agent provides execution with taste and discipline.

shareai-lab-kode Bundle

Guide for creating effective skills. Use when creating a new skill (or updating an existing skill) that extends an agent’s capabilities with specialized knowledge, workflows, or tool integrations.

shareai-lab-kode Skill

Troubleshoot Kode permission prompts/denials (tool allowlists, commandAllowedTools, dontAsk fail-closed, subagent inheritance). Use when tools are unexpectedly blocked, permission prompts repeat, or behavior differs between main agent and subagents.

shareai-lab-kode Skill

Kode capability-management playbook: treat features (LSP, statusline, output styles, plugins, notifications) as configurable capabilities and drive changes through the agent CLI (SlashCommand/Task) instead of pushing users into menu-like flows. Use when the user asks to enable/disable/configure a capability, diagnose why it’s off, or wants the agent to self-manage Kode’s own features (meta capability management).

endlessblink Skill

Create, validate, and publish Claude Code plugins and marketplaces. Use this skill when building plugins with commands, agents, hooks, MCP servers, or skills.

sundial-org Bundle

Fine-tune LLMs using the Tinker API. Covers supervised fine-tuning, reinforcement learning, LoRA training, multimodal models including Thinking Machines' Inkling (vision, audio, thinking effort), and both high-level Cookbook patterns and low-level API usage.

sundial-org Bundle

Calculate training costs for Tinker fine-tuning jobs. Use when estimating costs for Tinker LLM training, counting tokens in datasets, or comparing Tinker model training prices. Tokenizes datasets using the correct model tokenizer and provides accurate cost estimates.