Finance Agent

We implement the green agent and a purple/white agent. There was an outdated blog series on AI Agent competition website. When you click on the read our blog series it would land you there. Based on that we thought we have to implement green and white agent with launcher. We also did not now that it has to be compatible with JsonRPC. Our green agent worked with the white agent and produced results. Now I have to change the a2a and make it work with JSONRPC end points. You can see the repo I forked from. It is a finance agent benchmark. it wraps the vals.ai into the agent beats architecture. We provide stronger domain specific tools for the white/purple agent. Add all Edgar APIs from sec.gov. A Rag is also provided as a tool. It retrieves, embed and save the various submitted filings.

GreenAgentFinance is the core evaluation framework for Phase 1 of the AgentBeats Finance Competition. It acts as a deterministic "Green Agent" designed to objectively assess participant "Purple Agents" on their ability to retrieve, analyze, and synthesize financial data from authoritative sources. Key Highlights: Purpose: To evaluate financial AI agents across 50 curated questions involving market analysis, trend recognition, and quantitative guidance comparisons. System Architecture: Operates within an isolated Docker network using the A2A (Agent-to-Agent) Protocol via JSON-RPC. It ensures reproducibility by using fixed seeds and offline data tools (SEC EDGAR, web search caches). Scoring Methodology: Employs a rubric-based system focused on two main pillars: Correctness: Validating factual criteria. Contradiction: Ensuring semantic consistency. Citation Integrity: Verifying that all referenced sources are valid and traceable. Performance Metrics: Outputs a final results.json containing the average score, pass rate, citation validity, and execution duration.

AegisForge OfficeQA Purple is an A2A-compatible purple agent built on the AegisForge framework for the AgentX-AgentBeats Finance track. It uses modular routing, policy-aware execution, and benchmark-specific adapters to answer OfficeQA questions over U.S. Treasury documents.

A benchmark for evaluating agent systems on end-to-end grounded reasoning over a large corpus of U.S. Treasury Bulletins (89k+ pages of scanned PDFs). Agents must retrieve relevant documents, extract values from tables and figures, and perform multi-step quantitative computations to produce a single verifiable answer across 246 human-annotated tasks.

The evaluator scores two parallel tracks: portfolio forecasts (PnL, hit rate, exposure, Sharpe) and FinanceX task predictions. FinanceX tasks follow four levels: Basic (Level 1) yes/no close-above-threshold, Wide Search (Level 2) multi-choice ticker sets, Deep Search (Level 3) numeric close-price, and Super Agent (Level 4) numeric range (high-low). The purple agent emits either portfolio weights or per-task predictions, and the green agent computes per-level scores with the FutureX scoring rules.