{ "count": 8, "philosophy": "Large, cleaned, agent-ready datasets compiled only from public-domain and openly-licensed sources. Every pack ships with a LICENSE.md and a PROVENANCE.json naming every upstream source, its licence, and its retrieval date. No scraped forums, no leaked corpora, no personal data, and nothing sourced from charities or nonprofits. Buy a pack once and you get both a signed download and ongoing API access; Founding Gold Card holders get every pack included.", "sourcing_policy": "https://halcyoncompute.com/packs/sourcing", "included_with_gold": true, "gold_upgrade": "https://halcyoncompute.com/gold", "how_to_buy": { "catalog": "curl https://halcyoncompute.com/api/public/v1/packs", "detail": "curl https://halcyoncompute.com/api/public/v1/packs/sec-edgar-filings", "checkout": "curl -X POST https://halcyoncompute.com/api/public/v1/packs/sec-edgar-filings/checkout -H \"authorization: Bearer $HALCYON_KEY\"", "note": "Checkout returns a checkout_url. Present it to your human operator for approval — Halcyon does not ask agents to spend without consent.", "download": "curl https://halcyoncompute.com/api/public/v1/packs/sec-edgar-filings/download -H \"authorization: Bearer $HALCYON_KEY\"" }, "packs": [ { "slug": "sec-edgar-filings", "title": "SEC EDGAR Filings", "subtitle": "Every public-company filing since 1993, parsed and chunked.", "job": "Financial research, diligence, and compliance agents.", "tier": "A", "license": "CC0-1.0", "license_note": "Works of the United States government are not subject to copyright. No restrictions on commercial use.", "source": { "name": "U.S. Securities and Exchange Commission — EDGAR bulk data", "url": "https://www.sec.gov/os/accessing-edgar-data" }, "records": 21400000, "records_human": "21M", "size_bytes": 412000000000, "size_human": "412 GB", "formats": [ "jsonl", "parquet" ], "price_usd": 149, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/sec-edgar-filings", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/sec-edgar-filings/checkout", "human_page": "https://halcyoncompute.com/packs/sec-edgar-filings" }, { "slug": "us-federal-court-opinions", "title": "US Federal Court Opinions", "subtitle": "Roughly ten million opinions, Supreme Court down to district.", "job": "Legal research, citation graphs, and contract reasoning.", "tier": "A", "license": "CC0-1.0", "license_note": "Judicial opinions are in the public domain under the government edicts doctrine. Commercial use unrestricted.", "source": { "name": "CourtListener / Free Law Project bulk data", "url": "https://www.courtlistener.com/help/api/bulk-data/" }, "records": 9800000, "records_human": "9.8M", "size_bytes": 168000000000, "size_human": "168 GB", "formats": [ "jsonl", "parquet" ], "price_usd": 129, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/us-federal-court-opinions", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/us-federal-court-opinions/checkout", "human_page": "https://halcyoncompute.com/packs/us-federal-court-opinions" }, { "slug": "us-patents-full-text", "title": "US Patents (Full Text)", "subtitle": "Every granted patent since 1976, claims separated from spec.", "job": "Prior-art search, IP diligence, and R&D landscape mapping.", "tier": "A", "license": "CC0-1.0", "license_note": "USPTO bulk data is a US government work and carries no copyright restriction.", "source": { "name": "USPTO Bulk Data Storage System", "url": "https://bulkdata.uspto.gov/" }, "records": 8600000, "records_human": "8.6M", "size_bytes": 340000000000, "size_human": "340 GB", "formats": [ "jsonl", "parquet" ], "price_usd": 129, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/us-patents-full-text", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/us-patents-full-text/checkout", "human_page": "https://halcyoncompute.com/packs/us-patents-full-text" }, { "slug": "eu-law-eurlex", "title": "EU Law (EUR-Lex)", "subtitle": "All EU legislation and case law, twenty-four languages, aligned.", "job": "GDPR, DSA, and AI Act compliance work across jurisdictions.", "tier": "A", "license": "CC-BY-4.0", "license_note": "Reusable under Commission Decision 2011/833/EU. Attribution to the European Union required; commercial reuse permitted.", "source": { "name": "EUR-Lex / Publications Office of the European Union", "url": "https://eur-lex.europa.eu/content/help/data-reuse/reuse-contents-eurlex-details.html" }, "records": 1240000, "records_human": "1.2M", "size_bytes": 96000000000, "size_human": "96 GB", "formats": [ "jsonl", "parquet" ], "price_usd": 99, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/eu-law-eurlex", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/eu-law-eurlex/checkout", "human_page": "https://halcyoncompute.com/packs/eu-law-eurlex" }, { "slug": "pubmed-central-oa", "title": "PubMed Central Open Access", "subtitle": "Five million biomedical papers, only the licences that permit resale.", "job": "Clinical, pharmacological, and biomedical research agents.", "tier": "A", "license": "CC-BY-4.0", "license_note": "Commercial-use subset only. Articles under CC BY-NC or CC BY-ND are excluded from this pack by design.", "source": { "name": "NCBI PubMed Central Open Access Subset", "url": "https://pmc.ncbi.nlm.nih.gov/tools/openftlist/" }, "records": 5100000, "records_human": "5.1M", "size_bytes": 187000000000, "size_human": "187 GB", "formats": [ "jsonl", "parquet" ], "price_usd": 119, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/pubmed-central-oa", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/pubmed-central-oa/checkout", "human_page": "https://halcyoncompute.com/packs/pubmed-central-oa" }, { "slug": "noaa-ghcn-daily", "title": "NOAA Global Weather (GHCN-Daily)", "subtitle": "A hundred thousand stations, two centuries of daily observations.", "job": "Insurance, agriculture, energy, and logistics modelling.", "tier": "A", "license": "CC0-1.0", "license_note": "NOAA data is a US government work, free of copyright and free of licence conditions.", "source": { "name": "NOAA National Centers for Environmental Information", "url": "https://www.ncei.noaa.gov/products/land-based-station/global-historical-climatology-network-daily" }, "records": 3100000000, "records_human": "3.1B", "size_bytes": 128000000000, "size_human": "128 GB", "formats": [ "parquet", "csv" ], "price_usd": 79, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/noaa-ghcn-daily", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/noaa-ghcn-daily/checkout", "human_page": "https://halcyoncompute.com/packs/noaa-ghcn-daily" }, { "slug": "stack-exchange-dumps", "title": "Stack Exchange Q&A", "subtitle": "Every question, answer, and comment across all sites.", "job": "Coding agents: error resolution and real API usage patterns.", "tier": "B", "license": "CC-BY-SA-4.0", "license_note": "Share-alike. This pack is redistributed under CC BY-SA 4.0; anything you publish that incorporates it must carry the same licence and attribute the original authors. Model weights trained on it are generally not considered a derivative work, but take your own advice on that.", "source": { "name": "Stack Exchange Data Dump (Internet Archive)", "url": "https://archive.org/details/stackexchange" }, "records": 62000000, "records_human": "62M", "size_bytes": 94000000000, "size_human": "94 GB", "formats": [ "jsonl", "parquet" ], "price_usd": 89, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/stack-exchange-dumps", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/stack-exchange-dumps/checkout", "human_page": "https://halcyoncompute.com/packs/stack-exchange-dumps" }, { "slug": "openstreetmap-extracts", "title": "OpenStreetMap Extracts", "subtitle": "The whole planet, normalised into queryable entities.", "job": "Geocoding, routing, and location intelligence without an API bill.", "tier": "B", "license": "ODbL-1.0", "license_note": "Open Database Licence. You may use it commercially, but any derived database you publish must be shared under ODbL and attributed to OpenStreetMap contributors.", "source": { "name": "OpenStreetMap planet dump", "url": "https://planet.openstreetmap.org/" }, "records": 9200000000, "records_human": "9.2B", "size_bytes": 780000000000, "size_human": "780 GB", "formats": [ "parquet", "geojson" ], "price_usd": 99, "status": "preparing", "deliverable": false, "detail_url": "https://halcyoncompute.com/api/public/v1/packs/openstreetmap-extracts", "checkout_url": "https://halcyoncompute.com/api/public/v1/packs/openstreetmap-extracts/checkout", "human_page": "https://halcyoncompute.com/packs/openstreetmap-extracts" } ] }