From quaks-agents
Generates an investor briefing or answers financial questions using the Quaks MCP server. Replicates the News Analyst multi-agent workflow by loading prompts from the MCP server and executing them step-by-step. Invoke explicitly with /quaks-agents:news-analyst. Also use this skill when the user asks for a market briefing, daily investor report, financial news summary, stock news, market update, or wants to ask questions about recent market events, earnings, economic indicators, or investment topics — even if they don't mention 'quaks' or 'briefing' by name.
How this skill is triggered — by the user, by Claude, or both
Slash command
/quaks-agents:news-analystThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are replicating the Quaks News Analyst — a multi-step financial analysis workflow. This skill mirrors the LangGraph agent pipeline running on the Quaks backend by loading the same system prompts from the MCP server and executing each step sequentially.
You are replicating the Quaks News Analyst — a multi-step financial analysis workflow. This skill mirrors the LangGraph agent pipeline running on the Quaks backend by loading the same system prompts from the MCP server and executing each step sequentially.
The Quaks MCP server provides:
Prompts (loaded via prompts/get — these are the system prompts for each workflow step):
news_analyst_coordinator — Coordinator/QA mode system promptnews_analyst_aggregator — News aggregation system promptnews_analyst_reporter — Report writing system promptTools (called during workflow execution):
get_markets_news_mcp — Retrieves market news articles (used by aggregator step)get_insights_news_mcp — Retrieves AI-generated investor briefings (used by coordinator/QA step)Check whether the user supplied an argument after the skill name:
This mode answers the user's financial question using previously generated investor briefings as context. It mirrors what the coordinator node does when the query is NOT BATCH_ETL.
news_analyst_coordinator prompt from the MCP server.get_insights_news_mcp to fetch recent investor briefings. Use include_report_html=true if the question requires detailed analysis. Paginate with cursor if needed.Do NOT proceed to the aggregator or reporter steps. End here.
This mode generates a full investor briefing report. It mirrors the 3-node LangGraph pipeline: coordinator → aggregator → reporter. Execute each step sequentially — the output of each step feeds into the next.
The coordinator decides whether to proceed with briefing generation. In briefing mode, it always routes to the aggregator.
news_analyst_coordinator prompt from the MCP server.BATCH_ETL branch in the backend agent.The aggregator collects and prioritizes market news. This is the data-gathering step.
news_analyst_aggregator prompt from the MCP server.get_markets_news_mcp repeatedly to gather articles:
cursor to paginate through additional pages.search_term values to ensure broad coverage (e.g. "technology", "energy", "earnings", "federal reserve").The reporter produces the final polished briefing. This is the writing step.
news_analyst_reporter prompt from the MCP server.# Quaks Investor Briefing — [Today's Date]
> [One-sentence plain-language summary of the biggest theme today.]
## [Topic Headline 1]
[4 paragraphs]
## [Topic Headline 2]
[4 paragraphs]
...
---
*This automatically generated report is not equivalent to professional financial advice. Always do your own research before making any investment decisions. This report is not investment advice.*
*Quaks News Analyst — [Current Date and Time in UTC]*
These apply to the reporter output. They mirror the backend agent's reporter prompt:
npx claudepluginhub bsantanna/quaks --plugin quaks-agentsProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.