Research & Data

8个从 mattpocock-skills / anthropics-skills / addyosmani-agent-skills 提取的实用 skill（ask-matt, to-spec, mcp-builder, skill-creator, interview-me, matt-research, matt-triage, matt-wayfinder）

Agent skills for government economic, industry, policy, and brokerage-style research reports with deep research, data analysis, and visualization support.

Claude Code and Codex skill for compiler-exact .NET code search - references, callers and change impact, Razor and Blazor included. Roslyn-backed. A free tool by DBHQ.

Semantic catalog search with dynamic faceted attribute filtering (Anthropic Shopping)

Find the right problem. Make the next useful move.

Multi-product purchase research trade-off & spec analyzer (Anthropic Shopping)

Privacy peer-to-peer collaborative Markdown notes backed by GitHub. No server-side database.

Agent skills

Collection of my Claude Skills

Portable, js-compilable tools: schema validation, workflow routing, sqlite-querybuilder and soon others

Security-testing methodology as portable, tool-agnostic agent skills for Claude Code and any agent runtime: white-box bug hunting, AI-agent and LLM red-teaming, supply-chain risk, and appsec depth.

Curated AI skills for open data workflows — tool-agnostic instructions following the Agent Skills open standard

全流程 智能招投标 Agent：标书生成 · 招投标解读 · 标书检查 · 标书文档ai排版 · 商机发现 一键完成。 21 项合规检查 · 多模型切换 · RAG 知识库 · OCR 抽取。 从招标公告到可交付 docx 文档，全流程 AI 自动化。

Obsidian Vault Assistant is a local-first Codex skill and MCP server for people who want a searchable, maintainable Markdown knowledge base. It discovers configured Obsidian vaults, searches a three-layer model (Raw, evidence/middle, and Wiki), reads only the relevant chapter, and stages every change for explicit review.

本地 Agent Skill 管家:盘点全部本地 skill 的功能/来源/配套客户端,体检重复加载与链接漂移,生成 Markdown+HTML 报告,支持 --json 自动巡检 | Local agent-skills auditor with interactive HTML reports

Agent Skills for Neon Severless Postgres

A consent-first Codex skill for daily work reports and Feishu calendar planning

DVERA agent skills for CT C/C++ verification workflows: project setup, analysis, test generation, coverage, reporting, and regression.

Claude Skill for reviewing requirements/user stories against 7 quality criteria before test analysis — quote-based findings, severity levels, and open questions to the BA

Claude Code skills for running models cost-effectively: security gates (secret scanner, commit-gate), model-quality tooling (eval harness, token-squeeze, compact-io, opus-workers), review/advisory (trusted-advisor, audit, skill-vet, research-brief), and a read-only reorg-proposal advisor.

Evidence-grounded Codex Skill for protected data, controlled deformations, and reverse consistency.

Vet MCP and AI-agent tools before attachment. Local, deterministic static scanning with no tool execution, plus CI and redacted public records.

Generate private, reproducible Local Services Ads responsiveness reports from any raw-GAQL-capable Google Ads connector.

Brand Soul — an open-source framework created by Cacao Web Studio for persistent, evidence-aware AI brand identity.