{
  "markdown": "# HelixPay Company Knowledgebase\n\nThis repository contains a source-backed knowledgebase for a fictional B2B payments\ncompany, **HelixPay**.\n\nIt has two parts:\n\n- `data/` — the original messy company context dump.\n- `knowledge/` — a curated, agent-readable knowledgebase compiled from that dump.\n\nAn additional self-contained [Data Analytics MCP UI demo](examples/data-analytics-demo/README.md)\nshows how Codex can turn synthetic product-growth data into an interactive,\nsource-backed report with KPI cards, charts, and a sortable evidence table.\n\nThere is no app server, API, package install, or runtime required. The deliverable is\nthe knowledgebase itself.\n\n## What Is Included\n\nThe checked-in knowledgebase covers:\n\n- 44 source documents\n- 27 resolved entities\n- 31 sourced claims\n- 15 relationship edges\n- 10 scoped metrics\n\nThe structured source of truth is:\n\n```text\nknowledge/compiled/ontology.json\n```\n\nThe markdown knowledgebase is organized by concept type:\n\n```text\nknowledge/people/\nknowledge/teams/\nknowledge/products/\nknowledge/projects/\nknowledge/customers/\nknowledge/concepts/\n```\n\nStart with:\n\n```text\nknowledge/index.md\n```\n\n## What The Knowledgebase Demonstrates\n\nThe knowledgebase is intentionally built around the hard parts of company context:\n\n- **Entity resolution** — Maria Santos vs. Maria Silva, Daniel Tan vs. Tan Wei Ming,\n  and HelixPay POS vs. POS Self-Service.\n- **Contradictions and stale information** — public June messaging for Project\n  Confluence vs. the internal/board Q3 timeline; Q3 vs. Q4 timing for company-brain\n  platform value.\n- **Scoped metrics** — Q1 revenue, Q2 pacing gap, NPS by segment, Brazil SMB ARR at\n  risk, affected merchants, and churn impact.\n- **Causal chains** — Tap + Loyalty reconciliation defect, Açaí Express risk,\n  Cosmos Hotels churn, and multi-property reporting as an enterprise blocker.\n- **Org and ownership context** — reporting lines, ownership, blockers, and\n  dependencies.\n\nEach concept file includes citations back to the original source documents.\n\n## Layout\n\n```text\ndata/\n  Raw company corpus: docs, chats, emails, interviews, dashboards, PDFs, and images.\n\nknowledge/\n  Curated knowledgebase.\n\nknowledge/compiled/ontology.json\n  Machine-readable ontology containing documents, entities, claims, edges, and metrics.\n\nknowledge/index.md\n  Human/agent entrypoint for browsing the markdown knowledgebase.\n\nknowledge/log.md\n  Compilation log and superseded fact notes.\n\nexamples/data-analytics-demo/\n  Synthetic Data Analytics MCP report, executed notebook, and reproducible artifact builder.\n```\n\n## Source Corpus\n\nThe raw `data/` directory contains:\n\n```text\noverview.md\norg-chart.md\nall-hands-2026-04-15.md\nweekly-review-2026-04-21.md\nboard-update-2026-04-22.md\nq1-2026-results.pdf\nboard-deck-q1-2026.pdf\ninterviews/\ndashboards/\nimages/\nchat/\nemail/\ncode/\n```\n\nThe corpus is deliberately realistic: multiple formats, inconsistent naming, mixed\nlanguages, stale documents beside fresher ones, informal Slack alongside board\nmaterials, and contradictions that should be preserved rather than flattened.\n\n## Notes\n\n- All people, companies, and events are fictional.\n- The knowledgebase is checked in and usable as files.\n- No API key is required.\n- No build step is required.\n",
  "bytes": 3256,
  "sha": "6efc999541bb1e76300c27105886105db7f01c0640adf79bc9686faa3b6876f6",
  "repo_slug": "vood/company-brain-demo",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/okf_vood_company_brain_demo_knowledge_index__29293d8a/readme"
}