Research & Data

Portable Codex skills for controlled audit data analytics workflows · KPMG 审计数据分析技能包

Builds the current EVE Online Static Data Export into a compact, indexed SQLite database — small enough to upload anywhere — plus a Claude skill documenting the schema and the traps in the data.

Agent Skills for Linkly AI — search, browse, and read your local documents from any AI coding agent.

Codex skill for generating weekly work reports from recent AI work history

CLI and TypeScript client for Pontx public API discovery, OpenAPI Endpoint and Schema search, previews, and SDK integration

Local-first research mock interview system with pluggable LLM backend. / 科研模拟面试：本地优先，LLM后端可插拔（保研/申博/答辩）

Agent Skill: report per-conversation token usage, cache hit rate and estimated cost (DeepSeek peak/off-peak pricing) for ZCode, Codex, Claude Code, OpenCode

Local-first visual compiler and review workbench for Agent-managed Markdown knowledge

🏵️ MCP Server with CLI interface ready for distribution

DeepSeek Harness (DSH) 生态技能包：parallel-dispatch 并行调度规则 + archmap 架构测绘 Agent（零LLM确定性diff影响面，节约tokens）。Reusable skills for the DeepSeek Harness ecosystem.

AppLovin creative set 批量创建 Claude Skill — 视频+试玩分组建组、挂campaign、设CPP、默认PAUSED

Daily AI model radar: track, compare, and visualize model quality across domains. / AI模型雷达：每日追踪对比各领域模型质量与价格

A reusable AI-agent skill for Gulf and Middle East risk analysis across Iran sanctions, GCC banking, energy, and maritime chokepoints.

Multi-repo workflow isolation skill for DeepSeek Harness - one isolated entry agent per repo, one controller for cross-project dispatch, model routing & hash-chained ledger

开源 skill：把「首席科学家 + flash 研究员」团队协作模式固化为可复用规则手册——主模型只做规划/审查/汇报，便宜的小模型并行执行，省 pro 开销、省时间。

基金副驾：AI agent skill that screens Chinese mutual funds from public data for manual purchase through Alipay. Copilot, not autopilot — 它筛选，你决定。不登录、不下单、不预测。

Skills, prompts, and instructions for building AI agents on top of Dynatrace production context

Codex skill that locally scans files and text for C2PA, AI metadata and Unicode provenance signals

Agent skill for codebase analysis, requirement alignment, explicit approval, and style-consistent development

Grade Agent Skills against a production-grade quality rubric (3 hard gates + six-dimension weighted scoring). Evidence-based LLM-as-judge reports.

Binary-only static reverse-engineering CLI and agent Skill for bounded, evidence-backed analysis

Codex skill for converting FileOrganization tasks into independently judgeable metadata.json files

AI-agent skill for migrating Matillion ETL pipelines to Databricks

Privacy-focused Codex skill for diagnosing Lithe Audio speaker dropouts, delays and network issues. Checks a local IP, Wi-Fi, DHCP, latency, packet loss and roaming; provides guided fixes, permission-based router support and redacted reports without exposing credentials or internal APIs.