[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"WLMSyaRniZ":3},"# EconCSLib\n\nEconCSLib is a Lean 4 project for checking results in Economics and\nComputation. The repository has two roles:\n\n- Build a reusable library of mathematical tools for EC: probability,\n  optimization, matching, auctions, online algorithms, fair division, learning,\n  and related foundations.\n- Keep a paper-by-paper audit trail showing which source definitions and\n  theorems have been formalized, which assumptions remain, and where the proof\n  deviates from an informal paper argument.\n\nThe project is meant to support both formalization work and human review. A\nhuman reader should be able to open a completed paper folder and understand\nwhat was proved without reading the full Lean implementation.\n\nLinks:\n- [Project website](https://gargnikhil.com/EconCSLib/)\n- [Paper describing project](https://arxiv.org/abs/2606.13306)\n\n## How The Repository Is Organized\n\n- `EconCSLib/` is the reusable library. Code here should be paper-independent\n  and useful across more than one formalization.\n- `papers/` contains one folder per source paper. These folders preserve the\n  paper's notation, theorem numbering, proof DAG, validation report, and\n  human-facing Lean interface.\n- `docs/` contains project documentation. Some files are human-facing strategy\n  and status documents; others are detailed conventions for agents and\n  maintainers.\n- `skills/econcs-formalizer/` contains the agent workflow instructions used to\n  formalize papers consistently.\n\n## Reviewing A Formalized Paper\n\nStart in the paper folder under `papers/\u003CPaperName>/`.\n\nFor a completed or nearly completed paper, read these files in this order:\n\n1. `FINAL_VALIDATION_REPORT.md`: human-facing report summarizing deviations, assumptions, and final status.\n2. `PaperInterface.lean`: readable definitions and theorem statements matching\n   the paper. This is the main human-facing Lean file.\n3. Dependency graph: visual map of named definitions, lemmas, theorems, and\n   remaining caveats. Some paper folders keep only the source graph tracked and\n   render the PDF locally.\n\nImplementation-level proof files are for maintainers and agents. They should\nnot be necessary for a first human audit of what the paper claims and what Lean\nproves.\n\n## Current Status\n\nSee [`site/index.html`](site/index.html). Each paper folder also has a paper-local\n`status.json`.\n\nPaper IDs and folder names are stable artifact identifiers and may track an\narXiv, conference, or original working-paper year. Public status tables use the\npublished citation title and year.\n\n## Starting A New Paper With An Agent\n\nTo get started in formalizing your own paper, clone the repository and open an\nLLM agent tool (I use Codex with GPT 5.5 in xhigh thinking mode). Give the\nagent the paper link, and ask it to formalize the paper using the skill and\nworkflow in the repository. (And please let me know what your experience is\nlike!).\n\n## Development\n\nThis project is aligned to Lean/mathlib/CSLib `v4.30.0-rc2`.\n\nUseful commands:\n\n```bash\nlake build EconCSLib\npython3 scripts/audit_repository.py\n```\n\n`lake build EconCSLib` is the first fresh-clone check and should pass for the\npublic repository. `python3 scripts/audit_repository.py` is a maintainer audit.\nIn a fresh clone it may report missing ignored local artifacts such as source\nPDFs, rendered dependency-graph PDFs, or review-dashboard caches; those are not\nLean verification failures.\n\n## License\n\nUnless otherwise noted, the Lean source, scripts, documentation, and site source\nare licensed under the Apache License, Version 2.0. See [`LICENSE`](LICENSE).\nSource-paper PDFs and extracted text caches are not included in the public\nrepository unless redistribution rights have been checked separately.\n",1782661968380]