agentsclimarketplace

Query dma rag

Skill tinhct/lux-agent/.agents/skills/query_dma_rag

A multi-agent system for reliable algorithmic auditing and explainable regulatory compliance.

Install
npx -y skills add tinhct/lux-agent --skill query_dma_rag

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Searches indexed Digital Markets Act (DMA) documents via Vertex AI Search to retrieve precise legal definitions and compliance constraints regarding algorithmic self-preferencing. Use this skill when the user asks to cross-reference findings with the DMA, needs to check gatekeeper obligations, or requests the legal definition of self-preferencing. Do NOT use for searching the open web or providing binding legal advice.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.1 KB, as published. Nobody here has run it

Query DMA RAG

When to use

  • Cross-referencing technical data (e.g., scraped Amazon search results) against EU Digital Markets Act (DMA) gatekeeper rules.
  • Retrieving precise legal definitions of "self-preferencing," "core platform services," or "gatekeeper" from official regulatory texts.
  • Drafting the regulatory analysis portion of a compliance report that requires exact legal citations.

When NOT to use

  • Conducting general open-web searches or querying regulatory frameworks outside the indexed knowledge base (e.g., US antitrust law, unless explicitly indexed).
  • Generating definitive legal rulings, corporate liability verdicts, or binding legal counsel.

Workflow

  1. Receive the target legal concept or analysis parameter from the state graph (e.g., "DMA Article 6 rules on ranking").
  2. Execute a semantic search query against the Vertex AI Vector Search endpoint containing the indexed DMA documents.
  3. Retrieve the top-K relevant text chunks, isolating the associated article numbers, paragraphs, and document metadata.
  4. See references/rag_citation_guidelines.md for handling edge cases where retrieved chunks contain ambiguous language or cross-reference other un-indexed EU directives.

Examples

  • Input: "Define self-preferencing under the DMA." → Output: "Under the DMA, self-preferencing occurs when a gatekeeper treats its own services or products more favorably in ranking and related indexing and crawling than similar third-party services. (Source: DMA Article 6(5))"
  • Input: "Are search engines covered by gatekeeper rules?" → Output: "Yes, online search engines are defined as 'core platform services' subject to gatekeeper obligations if they meet the quantitative thresholds. (Source: DMA Article 2(2)(b))"

Output format

  • Use assets/regulatory_citation_template.md to format the response.
  • Always include the exact legal Article, Paragraph, and source document name directly alongside the extracted text.
  • You must automatically append the immutable disclaimer footer: ***Disclaimer: This analysis is for research purposes only. This is NOT legal advice.***

Anti-patterns to avoid

  • Don't hallucinate or heavily paraphrase legal definitions; rely on direct quotes from the Vertex AI Search chunks whenever possible.
  • Don't omit the source citations (Article and Paragraph numbers) in the final output.
  • Don't state that a specific company is definitively guilty of a legal violation; use investigative, objective language (e.g., "This technical observation intersects with restrictions outlined in...").

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.