{"data":{"slug":"llmquant-quant-mind","name":"quant-mind","tagline":"Intelligent knowledge extraction and retrieval framework for quantitative finance","github_url":"https://github.com/LLMQuant/quant-mind","owner":"LLMQuant","repo":"quant-mind","owner_avatar_url":"https://avatars.githubusercontent.com/u/67217001?v=4","primary_language":"Python","stars":2964,"forks":481,"topics":["agent","context-engineering","data","harness-engineering","knowledge","llm","pipeline","quantitative-finance","quantitative-research","workflow"],"archived":false,"github_pushed_at":"2026-08-15T07:49:16+00:00","maintenance_label":"Steady","stars_delta_30d":578,"url":"https://www.graphcanon.com/tools/llmquant-quant-mind","markdown_url":"https://www.graphcanon.com/tools/llmquant-quant-mind.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/llmquant-quant-mind","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=llmquant-quant-mind","description":"QuantMind is an open source agent-native knowledge extraction and retrieval framework for quantitative finance.","homepage_url":"http://llmquantdata.com/","license":"MIT","open_issues":32,"watchers":32,"ai_summary":"QuantMind is designed to aid in the analysis of financial data through advanced knowledge extraction and retrieval techniques.","readme_excerpt":"### 🚀 Quick Start\n\n#### The agent path (recommended)\n\nQuantMind is meant to be opened, not imported. Point a coding agent at the checkout and describe the pipeline you want:\n\n```bash\ngit clone https://github.com/LLMQuant/quant-mind.git\ncd quant-mind && claude      # or: codex\n```\n\nThen, in the agent session:\n\n> \"Build me a source-first paper artifact for arXiv 1706.03762, then persist it and search the summary.\"\n\nThe agent reads the repo's contracts (`AGENTS.md`), loads the relevant `contexts/` pages, writes the pipeline, and runs `scripts/verify.sh` before it hands the change back.\n\n#### The library path\n\nQuantMind is still a normal Python package. We use [uv](https://github.com/astral-sh/uv) for package management.\n\n```bash\nuv venv && source .venv/bin/activate\nuv pip install -e .\n```\n\n`PaperFlow` refines one arXiv PDF into a self-contained paper artifact. The cfg **type** selects the knowledge shape (`PaperStructureCfg` → `PaperStructureTree`, `PaperSemanticCfg` → `PaperSemanticResult`). Bind a `PaperStructureCfg` to build a source-native **structure tree** — a hierarchy of page-cited nodes:\n\n```python\nimport asyncio\n\nfrom quantmind.configs import PaperStructureCfg\nfrom quantmind.configs.paper import ArxivIdentifier\nfrom quantmind.flows import PaperFlow\n\n\nasync def main() -> None:\n    flow = PaperFlow(PaperStructureCfg(model=\"gpt-5.6-luna\"))\n    tree = await flow.build(ArxivIdentifier(id=\"1706.03762v7\"))\n    print(tree.id, len(tree.nodes))\n\n\nasyncio.run(main())\n```\n\nPrefer the **semantic** shape — a page-aware chunk set plus one cited global summary you can embed and retrieve over? Bind a `PaperSemanticCfg` instead — same class, different cfg:\n\n```python\nimport asyncio\n\nfrom quantmind.configs import PaperSemanticCfg\nfrom quantmind.configs.paper import ArxivIdentifier\nfrom quantmind.flows import PaperFlow\n\n\nasync def main() -> None:\n    flow = PaperFlow(PaperSemanticCfg(model=\"gpt-5.6-luna\", chunk_size=512))\n    result = await flow.build(ArxivIdentifier(id=\"1706.03762v7\"))\n    print(result.global_summary.summary)\n    print(result.source_revision.id, result.chunk_set.id)\n\n\nasyncio.run(main())\n```\n\n#### Fan out a batch with `batch_run`\n\n```python\nimport asyncio\nfrom datetime import datetime, timedelta, timezone\n\nfrom quantmind.configs import NewsCollectionCfg, NewsWindow\nfrom quantmind.flows import batch_run, collect_news\n\n\nasync def main() -> None:\n    end = datetime.now(timezone.utc)\n    windows = [\n        NewsWindow(\n            source=\"pr-newswire\",\n            start=end - timedelta(days=day + 1),\n            end=end - timedelta(days=day),\n        )\n        for day in range(3)\n    ]\n    result = await batch_run(\n        collect_news,\n        windows,\n        cfg=NewsCollectionCfg(retain_raw_html=False),\n        concurrency=3,\n        on_error=\"skip\",\n        on_progress=lambda done, total: print(f\"{done}/{total}\"),\n    )\n    print(f\"ok={result.success_count} failed={result.failure_count}\")\n\n\nasyncio.run(main())\n```\n\n#### Resolve free-form intent with `magic`\n\n```python\nimport asyncio\n\nfrom quantmind.flows import collect_news\nfrom quantmind.magic import resolve_magic_input\n\n\nasync def main() -> None:\n    inp, cfg = await resolve_magic_input(\n        \"Collect the last day of PR Newswire company news.\",\n        target_flow=collect_news,\n    )\n    batch = await collect_news(inp, cfg=cfg)\n    print(f\"documents={batch.success_count} complete={batch.complete}\")\n\n\nasyncio.run(main())\n```\n\nMore examples live under [`examples/`](examples/); design contracts live under [`contexts/design/`](contexts/design/).\n\n---\n\n### License\n\nQuantMind is released under the MIT License—see `LICENSE` for details.","github_created_at":"2025-03-14T02:51:22+00:00","created_at":"2026-07-15T10:49:25.375384+00:00","updated_at":"2026-09-20T04:27:47.794291+00:00","categories":[{"slug":"data-retrieval","name":"Data & Retrieval","url":"https://www.graphcanon.com/categories/data-retrieval","markdown_url":"https://www.graphcanon.com/categories/data-retrieval.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/data-retrieval"}],"tags":[{"slug":"data","name":"data"},{"slug":"knowledge","name":"knowledge"},{"slug":"llm","name":"llm"},{"slug":"pipeline","name":"pipeline"},{"slug":"quantitative-finance","name":"quantitative-finance"},{"slug":"quantitative-research","name":"quantitative-research"},{"slug":"workflow","name":"workflow"}],"trust":{"provenance":{"is_fork":false,"github_id":948264786,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-17T06:00:31.035Z","maintenance":{"label":"Steady","score":60,"methodology":"github_public_v1","releases_90d":0,"days_since_push":32,"last_release_at":null,"stars_delta_30d":578,"open_issues_delta_30d":3},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-15T10:49:26.791Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-17T06:00:31.516Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-09-17T06:00:31.516Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-09-17T06:00:31.516Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["Use QuantMind when you need specialized tools for quantitative finance that can handle complex knowledge extraction and retrieval processes efficiently.","Ideal for environments where detailed market or historical financial data needs precise analysis using Python."],"when_not_to_use":["Avoid using QuantMind if your project does not involve quantitative finance, as its specific functionalities may offer limited value in non-finance areas.","Do not use this framework if you do not require advanced knowledge extraction and retrieval mechanisms or prefer simpler tools without integration with the uv package manager."],"source":"enrich:decision_facts","observed_at":"2026-07-17T08:22:25.367Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"QuantMind is an intelligent knowledge extraction and retrieval framework for quantitative finance, leveraging advanced techniques to assist in the analysis of financial data."}]}}