Back to directory

Asgard Skills

An open-source knowledge base of 301 coding-agent skills across 22 subject categories, from graduate-level theory frameworks and algorithms to Taiwan-specific business and tax knowledge, packaged as Claude Agent Skill Markdown files.

Asgard Skills

Asgard Skills packages 301 skills as independent Markdown files following the Claude Agent Skills format, each capturing the methodology, judgment calls, and pitfalls for one specific analytical task rather than generic advice, with some skills including plain Python scripts for deterministic calculations. The skills are grouped under 22 category prefixes, covering graduate-level theory models such as CAPM, structural equation modeling, and resource-based view, algorithms such as PageRank, BM25, and ARIMA, business-school frameworks such as SWOT and Porter's five forces, and localized Taiwan knowledge on e-commerce, the stock market, taxation, and electronic invoicing, alongside smaller categories for law, statistics, UX methodology, and general technical topics.

The repository is described as the raw-material library for the Asgard AI Platform, where these skills are combined with Asgard's own MCP servers and packaged into coding-agent plugins aimed at specific user roles, such as a stock analyst or an e-commerce operator.

Each skill's catalog entry gives a short description of what it does and when to use it, drawn from that skill's own frontmatter.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.