AI & ML

latitude-dev Bundle

PromptL syntax guide for writing prompts in the Latitude platform. This skill should be used when writing, reviewing, or editing PromptL prompts. Triggers on tasks involving creating prompts, configuring LLM parameters, using variables, conditionals, loops, chains, tools, agents, or any prompt engineering in Latitude.

latitude-dev Bundle

Integrate Latitude telemetry into any backend project for LLM observability. Use when the user wants to add Latitude tracing, logging, or observability to their LLM calls. Covers three approaches based on language and needs -- SDK telemetry wrapper (Python, TypeScript), raw OpenTelemetry OTLP (any language with OTel), or the Create Log API (any language with HTTP). Triggers on tasks involving Latitude telemetry, tracing LLM calls, sending traces to Latitude, or adding LLM observability.

learnprompt Bundle

管家阿福(Afu)——把 Obsidian 收件箱变成活的 Wiki,再变成本周待办卡和周历:Inbox -> Wiki -> Todo Card -> Calendar 一条线。本地优先,Markdown 是唯一真源,收件箱原文不改写,导入/排期/作废/编译全部写操作日志可回滚。 触发词包括但不限于:整理我的 Obsidian 收件箱、每天提醒我处理 Obsidian 收件箱、inbox 清零、把这批资料编译成 Wiki、从笔记生成本周待办、待办卡排进周历、阿福帮我理一下 vault。 即使用户只是说"我的 Obsidian 堆了一堆没处理的东西",也应该触发。 不要用于:不涉及 Obsidian/Markdown vault 的普通待办管理(用一般 todo 工具);改写、润色、总结笔记原文(阿福只整理位置,不动原文);Notion 等非 Markdown 知识库。

learnprompt Bundle

Loop Engineering（循环工程 / 环路工程）落地脚手架，人格代号「愚公」。把一句模糊的「我想自动搞定 X」，装配成 goal、intake、trigger、worktree、maker/checker、connectors、state、verification、guardrails 都齐备的自跑 Loop，最后把可一键执行的命令交到你手上——扳机你自己扣。 这是 Prompt Engineering → Context Engineering → Harness Engineering 之后的第四阶段 Loop Engineering 的工程化落地：你不再一轮一轮手动提示 Agent，而是设计一个能自主发现任务、执行、验证、迭代到目标达成的循环。产出一份组件矩阵、一份 runtime 中立的《Loop Spec》、Claude Code 与 Codex 两套可直接运行的命令、以及 manifest/日志等状态文件。 触发词包括但不限于：做一个 Loop Engineering、把这个任务做成 loop、把任务做成 goal、设计一个循环工程、loop engineering、循环工程、环路工程、让 agent 自己跑、自主工作流、把维护/triage/升级做成自动循环、帮我开 goal、设计 /goal、设计 /loop、跑一个能自己迭代的 agent、愚公。 即使用户只是描述「我想让 agent 每天/反复自动做某事，自己迭代到完成」，只要意图是把人工逐轮提示变成可自跑的循环系统，都应触发。 不要用于：从零写一个普通 skill（用 skill-creator）、打磨/升级单个已有 skill（用鲁班 luban）、一次性的单轮任务（直接做即可，不必上 loop）、普通代码 review（用 code-review）。

learnprompt Bundle

One source of truth for local AI agent skills: audit Codex, Claude, Hermes, OpenClaw, OpenCode, workspace skills, and shared roots; score hygiene, diff conflicting installs, classify drift as pristine/dirty/local-only against recorded baselines, deduplicate into one canonical source with symlinks, and migrate the same layout across machines with restorable backups. Trigger when the user says things like 'sync my skills', '同步我的 skills', 'check skill drift', '检查 skill 漂移', 'dedupe my skills', 'skill 大扫除', 'audit my skills', 'which copy of this skill is canonical', or wants one source of truth across multiple agents. Do NOT trigger for installing a brand-new skill from a catalog or registry, for editing one skill's content, or for single-host questions that need no cross-host comparison.

learnprompt Bundle

搭建或运行通过 UU 远程 macOS CLI 采集多台 Mac 上 Codex Agent 状态的每日晨报，包含隐私 Hook、终端安全采集、任务进度判断、日报图和定时任务。适用于 UU 远程 × Codex 日报、夜间 Agent 汇总和多设备进度巡检。

kortix-ai Bundle

Full semantic search over Desktop files, agent memory, and knowledge. Use when the agent needs to find relevant files or search knowledge semantically.

leeguooooo Skill

Use when a task needs a real iPhone — operating iOS apps that have no API (Apple Health, banking, IM apps), exporting on-phone data, tapping/typing/scrolling on the phone, or taking phone screenshots. Drives the iphone-use daemon's Direct/WDA HTTP agent API.

leeguooooo Skill

Read, search, send, and manage email across Gmail, QQ, 163, Outlook and any IMAP/SMTP account from the command line. Use when the user asks to "read my email", "查邮件", "look up an Amazon order email", "find the customer review notification", "send an email", "回复邮件", "delete spam", "查未读", "show unread", "synchronize my mailbox", "set up MCP for email", or anything that involves listing / searching / reading / writing / classifying messages from one or more mailboxes.

leeguooooo Skill

Talk to any other AI coding agent on this machine (Claude Code sessions, Codex tasks, Pi sessions, terminal TUIs) over open-cross-session. Use when asked to discuss with, delegate to, wake, or message another local agent/session, or to check what other agents are running.

leeguooooo Bundle

CLI & agent skill for a PayPay証券 (PayPay Securities, ペイペイ証券) account: check portfolio, holdings, balance, 投資信託, 米国株, 取引履歴 (transaction history), fees & FX-spread; generate a 復盘/review with realized & unrealized P&L and money-weighted return (XIRR); 持仓结构/exposure (risk), 定投/つみたて recurring-buy plans (plans), per-year tax view (tax), account snapshots & diff (snapshot/diff), multi-account consolidation (-a all), Chinese output (--lang zh); AND place 米国株 orders (buy/sell/cancel). Use when the user wants to view, review, analyze, report, or trade a PayPay証券 / PayPay investment account. Orders dry-run by default; a live order requires an explicit human `--execute` with a typed confirmation + the account TRADE_PASSWORD (the agent never auto-submits). Facts only — never investment advice or buy/sell judgments.

leeguooooo Bundle

macOS WeChat CLI + local HTTP bridge + Wechaty Puppet gRPC gateway — send messages, query sessions / contacts / chat history / images / favorites, and expose stable HTTP / gRPC surfaces for agent integration. Use when the user asks to 'send a WeChat message', '发微信', query WeChat contacts/groups/messages, look up who said what in a chat, fetch images from history, export chat history, wire WeChat into Hermes / n8n / Dify / LangChain, or run any wechaty bot on a real macOS WeChat account. Uses the installer-managed, isolated WeChat 4.1.9 clone on macOS (Apple Silicon) and a `wechatuse_` activation code. One-time `wechat-use init` extracts the DB key; no sudo, no re-signing WeChat.app. Optional remote bridge — `wechat-use tunnel setup --hostname YOUR_HOSTNAME` exposes the local REST API via Cloudflare Tunnel for remote services to call.

leeguooooo Skill

DEPRECATED — prefer the official zentao-cli (https://github.com/easysoft/zentao-cli) and its skills (https://github.com/easysoft/zentao-skills) for new setups. This skill still works for existing @leeguoo/zentao-mcp users and remains useful for its bug resolution stats command. Use the zentao CLI to query and operate ZenTao bugs, tasks, stories, todos, products, programs, projects, executions, plans, releases, test cases, test tasks, test suites, docs, users, departments, issues, and risks. Use when the user mentions 禅道 or ZenTao, wants bug/task/story/todo/project/test/doc lookups or updates, wants bug resolution rate stats or reports (bug 解决率统计/解决数量统计/bug statistics/bug report) by product or person, or needs login / whoami / self-test for a 禅道 instance. ZENTAO_URL usually includes /zentao.

kulaxyz Bundle

Cut token usage across the whole session without losing correctness or substance — concise user replies, lean tests/docs/plans/comments, tight code, disciplined context and tool use, and cheap sub-agent delegation. Load when the user wants less verbosity, lower cost, fewer tokens, or a "token diet" / "lean mode"; also triggered by /token-diet or "token diet on".

kunchenguid Skill

Use acpx as a headless ACP CLI for agent-to-agent communication, including prompt/exec/sessions workflows, session scoping, queueing, permissions, and output formats.

letta-ai Bundle

Contributes an agent to the letta-ai/agent-file directory. Use when setting up a new agent (.af file) for publication, including directory creation, avatar processing, file placement, and structure verification. Triggers on requests to release, publish, or contribute an agent file.

letta-ai Skill

Develops and tests the letta-ai/letta-acp adapter. Use when changing ACP protocol handling, Letta Agent SDK integration, sessions, streaming, tool calls, approvals, cancellation, editor filesystem delegation, slash commands, backends, acpx compatibility, packaging, or CI in this repository.

kunchenguid Skill

Use when the user asks to run GNHF, says they are going to sleep or leaving and wants an agent-managed coding run, asks to supervise, steer, or review an active GNHF run, or gives feedback on GNHF results.

momozi1996 Bundle

天涯神贴大神智囊团 — 20位传奇大神组成的多Agent决策系统。 基于ClawTeam多Agent协作原理，实现「一个目标 → 20位大神并行分析 → 综合决策」的完整工作流。 核心机制： - 智能领导者Agent（协调者）接收目标，分配给20位专业大神 - 20位大神并行分析，各自输出专业视角 - 综合决策Agent整合所有视角，输出最终决策 使用方式： - 「启动天涯智囊团」+ 你的问题/目标 - 「召集大神」+ 你的问题/目标 - 「20位大神分析」+ 你的问题/目标 涵盖领域：权谋分析、经济预言、历史解读、文化解构、都市情感、技术分析、 黑帮江湖、盗墓探险、玄幻武侠、都市异能、悬疑推理、法医刑侦、历史穿越等。

mvanhorn Bundle

Install agentcookie on a Mac source and a Linux or Mac sink so Chrome cookies sync continuously over Tailscale. Use when the user says "install agentcookie", "set up cookie sync", "share my Chrome sessions with my agent box", or "make my agent log in as me".

mvanhorn Skill

Curate production-grade system prompts for autonomous coding agents and multi-agent orchestration. Use when creating, improving, or debugging agent prompts. Not for user-facing messages or chat prompts.

acedergren Skill

Generate an Agent Workflow Roast report from recent local session history and optional memory context. Use when the user invokes @roast, /roast, or asks for workflow coaching with receipts.

acedergren Skill

Use when a Codex session should recall, write, seed, inspect, or delete curated Oracle-backed memory through the oracle-codex-memory MCP sidecar.

acnlabs Bundle

Agent Collaboration Network — Register your agent, discover other agents by skill, route messages, invoke another ACN agent through AgentRouter, manage subnets/orgs, work on Org work items or Task Pool tasks, register an execution workspace doorplate (acn workspace), and connect yourself to Interfaze chat (Mode A direct or Mode B listen+writeback) when the user wants to talk on interfaze.io. Use when joining ACN, finding collaborators, sending or broadcasting messages, calling an agent by id (AgentRouter), Org Harness (acn org), execution workspace / 进场, accepting and completing assignments, or enabling Interfaze / AgentPlanet chat.

acnlabs Bundle

Distill any commercial entity into a personalized brand agent — a living brand persona with authentic voice, declared service capabilities, and a standard service contract. Every commercial entity has a brand: a name, a style, a way of showing up in the world. This skill exists so that a street vendor, a family clinic, and a global chain can all have their own agent on equal footing. Supports both distillation from existing brand content and declaration from scratch.

acnlabs Bundle

Persistent, incremental, searchable persona knowledge base. Ingests data from Obsidian vaults, chat exports, X/Twitter archives, and more into a MemPalace-backed store with a Karpathy LLM Wiki knowledge layer. Exports training/ directories for persona-model-trainer.

addyosmani Bundle

Multi-agent adversarial code review. Spawns three reviewer subagents (Auditor, Adversary, Pragmatist) on a single model, runs a cross-examination round, then deterministically synthesizes a single ranked report. Use this for non-trivial PRs, security-sensitive changes, refactors that touch many files, or any change where one perspective has obvious blind spots. Trigger phrases include "adverse review", "adversarial review", "multi-perspective review", "review my changes from multiple angles", "panel review". NOT suitable for trivial diffs (typos, dependency bumps, formatting).

agent-next Bundle

Best practices for setting up high-quality GitHub repos for AI coding agents. Use when setting up a new repo, improving an existing repo's infrastructure, or answering "what does this repo need for agents to work effectively". Triggers on "set up repo", "make repo agent-ready", "repo best practices", "/agent-ready".

agentenatalie Bundle

Evidence-based cat behavior consultation skill for cats and companion animals. Use when the user asks for help with cat aggression, fear, stress, anxiety, elimination problems, pain- or disease-related behavior change, clinic handling, welfare, behavior triage, or a cat behavior consultation. Guides intake, retrieves local literature with bundled scripts, can use Zotero MCP or optional PaperQA2, and produces cited consult reports with medical-first triage.

ahmadawais Skill

Browse, filter, and compare AI models from models.dev in the terminal. Use when the user needs to find model IDs, inspect providers, compare context windows or pricing, filter by capabilities or status, or get machine-readable model metadata for scripts and agents.

automatelab-tech Bundle

Notion-as-source-of-truth dispatch board. Status flow Suggestion→Discussion→To-Do→In Progress→Done with subtasks, recurring tasks, deps, batch agent execution. Trigger: 'suggest X', 'discuss X', 'approve', 'start X', 'run todo', 'mark X done', 'kill X', 'next', 'capture this to Notion', 'agency-os status'.

0sxd Skill

Teaches agents how to read, query, and contribute to the corpus inventory wiki. Provides the classification taxonomy, wiki page contracts, and update protocols for maintaining the multi-agent knowledge index.

0antique Bundle

Help Codex plan, generate, reproduce, and refine editable Microsoft Visio scientific diagrams. Use when users ask Codex to operate local Visio, create VSDX research figures, redraw uploaded paper screenshots or sketches in Visio, convert natural-language method descriptions into scientific diagrams, export preview PNG/PDF/SVG, or build editable model architecture, workflow, system architecture, experiment, pseudo-code, sequence, or module interaction diagrams.

0xmassi Bundle

Apply a workflow from the Astra prompt pack when the user asks for an Astra audit, planning, research, or implementation workflow, or selects a numbered task from the pack. Use the selected workflow within the user's stated scope; ordinary coding questions do not need the whole pack.

0xsero Skill

Ship the current work — repo, uncommitted changes, .env files, and the live agent session — to another machine over Tailscale SSH, where the agent resumes in tmux and keeps working. Use when the user says /parcel, "hand off to <machine>", "send this work to another machine", "keep this running after I close the laptop", or wants to move a coding session (Claude Code, Codex, pi, Droid, or Grok Build) to a remote box.

0xsero Bundle

Use Parchi Relay to control a real browser via the Parchi extension agent (WebSocket) or directly via JSON-RPC from a CLI/tooling workflow.