AI-powered multi-agent equity research in Python

I built Hermes, an open-source Python framework for multi-agent financial research.

Most AI “equity research” demos stop at generating text. In practice, real workflows require pulling structured XBRL financials from SEC filings, extracting labeled sections (MD&A, Risk Factors, etc.), merging macro and market data, building actual Excel models with formulas, and generating investment memos in Word or PDF.

Hermes is designed to handle that full pipeline.

It includes:

* 35 financial data tools (SEC EDGAR via edgartools, FRED, Yahoo Finance, RSS news) * Composable agents (filings, macro, market data, modeling, report generation, orchestrator) * Excel workbook generation via openpyxl * Word + PDF report generation * ChromaDB indexing for semantic search over filings * Async rate limiting and file-based caching (filings cached permanently, quotes never cached) * Streaming progress events

Everything is async and usable standalone — you don’t need to use agents if you just want the tools.

Example: generate a DCF + research report in one call:

``` from hermes import Hermes, configure

configure(sec_user_agent="HermesFinancial/0.1 ([email protected])") h = Hermes(model="gemini-3-flash-preview")

result = h.invoke("Build me a DCF and investment research report for GOOG") print(result["response"]) ```

You can also directly fetch multi-period XBRL company facts, extract structured financial statement tables, parse qualitative sections from filings, generate formatted Excel models with formulas and charts, and export investment reports to .docx or PDF.

It’s MIT licensed and designed to be extended. You can register custom tools and agents easily.

Curious to hear feedback from both AI engineers and finance folks, especially around validation, reliability, and real-world research workflows.

Repo: [https://github.com/schnetzlerjoe/hermes](https://github.com/schnetzlerjoe/hermes)