quant-mind
QuantMind is an agent-native knowledge extraction and retrieval framework for quantitative finance.
QuantMind is an agent-native knowledge extraction framework for quantitative finance that transforms raw financial documents—research papers, news, filings—into structured, typed knowledge artifacts. Designed as a repository-level harness, it enables coding agents to build sophisticated financial information pipelines through contracts, progressive-disclosure contexts, and portable skills. The framework processes sources deterministically into timestamped, citation-backed knowledge shapes (papers, news, earnings) that support retrieval-augmented generation and agentic research workflows.
Key Features
Use Cases
- 01Building financial research pipelines that extract structured insights from arXiv papers and academic publications
- 02Collecting and processing multi-day news batches from sources like PR Newswire with automated chunking and summarization
- 03Creating retrieval-augmented generation systems over typed financial knowledge with citation tracking
- 04Enabling coding agents to develop quantitative finance data workflows through natural language instructions
- 05Powering agentic research assistants that perform deep financial document analysis with temporal awareness
- 06Constructing semantic search libraries over financial papers with page-aware chunking and global summaries
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
quant-mind — FAQ
What is QuantMind and what does it do?+
QuantMind is an agent-native framework that refines raw financial information (papers, news, filings) into structured, typed knowledge artifacts with timestamps and citations. It's designed as a repository harness that coding agents open and work within, rather than a traditional Python library to import.
How do I install and use the QuantMind agent skill?+
Clone the repository and open it with a coding agent like Claude or Codex. The agent reads AGENTS.md contracts and contexts/ documentation to execute workflows. Alternatively, install as a Python library using uv: 'uv venv && source .venv/bin/activate && uv pip install -e .'.
Which AI agents and clients work with QuantMind?+
QuantMind provides portable skills mirrored for both Claude (via .claude/skills/) and Codex (via .agents/skills/). The quantmind-dev skill supports contributor setup, issue filing, commits, and PRs in both environments.
Do I need API keys or external services to use QuantMind?+
Yes, you'll need access to an LLM API (e.g., GPT models) for the extraction flows, as PaperFlow and other operations accept a model parameter like 'gpt-5.6-luna'. The deterministic preprocessing steps do not require model access.
Is QuantMind free and open source?+
Yes, QuantMind is released under the MIT License and available as open source on GitHub. The codebase is free to use; note that LLMQuant Data is a separate commercial product built on top of the open framework.
What Python version does QuantMind require?+
QuantMind requires Python 3.8 or later. The project uses uv for package management and asyncio for concurrent operations.
How do I install quant-mind?+
Open the source repository on GitHub and follow its README. quant-mind is a skill — MCP Agents Market links you directly to the official repo.
Is quant-mind free?+
quant-mind is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.