{
  "markdown": "# canlicapital.com\n\nThe public site for **Canli Capital**, and the surface where **ALPHAC** — a four-sleeve\ncross-asset quant book — publishes its record while it is still small enough to be embarrassing.\n\n**Created and maintained by [Arhan Canli](https://github.com/arhancanli) for Canli Capital.**\nDevelopment uses reviewed AI-assisted tooling, while project ownership, research decisions,\npublished claims, and release responsibility remain with Arhan Canli.\nMachine-readable software citation metadata is provided in [`CITATION.cff`](CITATION.cff).\n\nThe engine that produces every number here is open too:\n**[github.com/arhancanli/alphac](https://github.com/arhancanli/alphac)**.\n\nLive: **[canlicapital.com](https://canlicapital.com)**\n\n## Why this repo is public\n\nThe site's whole claim is *\"a quant fund proving itself in public before it asks you to trust\nit.\"* — the tagline in `config/brand.js`. A site that makes that claim and hides its own source\nis asking for a trust it hasn't earned. So: this is the source, including the parts that enforce\nhonesty on us.\n\nThe load-bearing one is **`docs/retracted_claims.txt` in the engine repo**. When a number is\nwithdrawn it goes on a blocklist, and `check_retracted_claims.py` scans `dist/` and `public/` for\nit before each deploy. It cannot be satisfied by deleting the number — a retracted figure must\nstill be quotable *inside its own retraction*, so a match counts only when the explanation is\nabsent from the surrounding window.\n\nThat check exists because the retraction here had already failed twice: a withdrawn DSR of 0.83\nstayed on the homepage and in the social-unfurl card for six days after the signed chain formally\nwithdrew it. The pipeline was publishing the correction and the error in the same run.\n\nThe gate is now fail-closed for publication. Since 2026-08-19, both `live_tick.sh` and\n`live_publish.sh` run `check_retracted_claims.py` after regeneration and skip the deploy when it\nfails. Trading remains outside that blast radius: a publication defect can stop the website from\nshipping, but cannot place, cancel, or delay an order.\n\n## Publication surfaces\n\n| surface | what it is |\n|---|---|\n| `index.html` | the landing: thesis, systems teaser, live record |\n| `systems.html` | how the four sleeves work |\n| `research.html` | the research programme, literature reviews, feasibility protocols |\n| `performance.html` | the methodology and the honest numbers |\n| `progress.html` | the build log |\n| `open.html` | proven in the open: the kill log, the signed chain, glass-box artifacts |\n| `verify.html` | independent verification instructions and downloadable evidence |\n| `review.html` | the governed public criticism bench for five flagship papers |\n| `foundry.html` | the fail-closed design and deployment-acceptance status for Foundry |\n| `founder.html` | the ProfilePage that resolves every Arhan Canli authorship claim |\n| `methodology.html` | evidence-linked answers to the research methodology questions |\n| `research/*.html` | 111 generated technical reports, each with Scholar metadata and BibTeX |\n| `research/topics/*.html` | 13 substantive subject and research-stage indexes |\n| `measurements/*.html` | 89 generated Dataset pages with explicit claim boundaries |\n| `engineering.html` | the open-source hub: the three repositories, what is hard in them, and a reading path |\n| `notes/*.html` | engineering notes: post-mortems, derivations and design arguments |\n| `tools/selection-risk.html` | the Selection Risk Lab: search a series with no edge, watch the deflation kill what you find |\n| `tools/breadth.html` | the Breadth Lab: what a book of N sleeves is worth, and the ceiling no amount of breadth can pass |\n| `tools/execution.html` | the Execution Reality Lab: which execution assumptions are costs, and which only look like costs |\n| `tools/backtest-overfitting.html` | probability of backtest overfitting by Combinatorially Symmetric Cross-Validation, run in the browser on your own matrix |\n| `tools.html` | the index of every browser calculator this project publishes |\n| `developers.html` | the public read API: endpoints, the response envelope, and what each response cannot be used to claim |\n| `costs.html` | every cost that can reach a return, whether the engine charges it, and which way the answer is wrong when it does not |\n| `standards/paper-evidence.html` | canli.paper-evidence.v0, a proposed open standard whose required fields are the ones a performance claim usually omits |\n\n`public/paper-state.json` and `public/glassbox/*` are written by the engine's publish job, not by\nhand. They are the machine-readable form of every claim the pages make. Current corpus counts are\nderived during the build from `public/research-index.json`,\n`public/glassbox/trial_packet_manifest.json`, and the generated measurement directory; the sitemap\nis generated from the same files rather than maintained separately. The present build contains\n263 canonical URLs in the sitemap (all indexable), plus a public noindex evidence page for every incomplete registered\ntrial and one archival HTML paper per registered sleeve. It publishes\nidentity-level packets for all 228 recorded hypotheses, while\nhonestly marking 226 of those packets incomplete.\n\n## Build and run\n\n```sh\nnpm install\nnpm run build      # Vite multi-page build -> dist/\nnpm run preview    # serve the built dist\nnpm run dev        # dev server with hot reload\n```\n\nThree.js, GSAP + ScrollTrigger and Lenis are self-hosted — they install from npm and Vite\nfingerprints them into `dist`. **Nothing is fetched from a CDN at runtime.** The build must stay\ngreen, and source plus `dist` must contain zero em dashes (U+2014); both are audited before deploy.\n\n## Brand and facts are single-sourced\n\n`config/brand.js` is the one place names and numbers live. `js/shell.js` renders the nav and footer\nfrom it so every page ships byte-identical chrome, and `js/main.js` binds `data-brand`,\n`data-flagship`, `data-tagline` and `data-fact` nodes from it.\n\n**`STATS` and `FACTS` are the only numeric claims permitted on the hand-authored marketing\nsurfaces.** Generated papers and measurement pages obtain their figures from engine exports and\ncarry their own source paths and claim boundaries. `audit-published-numbers.mjs` reconciles the\nshared site-level figures; do not add a number to either layer without binding it to an\nauthoritative artifact.\n\n## Not investment advice\n\nNothing on this site or in this repo is investment advice, an offer, or a solicitation. The record\npublished here is **paper trading**; the published ALPHAC strategy record includes no funded performance. Simulated and past\nperformance do not indicate future results. See `LICENSE`: provided \"as is\", without warranty.\n",
  "bytes": 6743,
  "sha": "9a15f751eb3eb96e56d86b59c9d6c1e06d0d8b493cd1f37b107445513cbe5839",
  "repo_slug": "arhancanli/canlicapital",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_arhancanli_canli_validation_mc_de969740/readme"
}