# Project export: NexusOS This document was generated by HackStack to give an AI agent context about a hackathon project. Sections are labeled with their provenance; content marked as truncated was cut to keep this document small. ## Project metadata - Hackathon: OpenAI Build Week - Tagline: An AI operating system that turns scattered workspace context into structured, explainable missions using GPT-5.6, specialized agents, and evidence-backed recommendations. - Devpost: https://devpost.com/software/nexusos-xghvqc - GitHub: https://github.com/ArnavMishra23/NexusOS - Video: https://www.youtube.com/embed/1XK-zGbsi=3XblM1Tivj7JYO69y?enablejsapi=1&hl=en_US&rel=0&start=&version=3&wmode=transparent - Team: 1 GitHub contributor(s) — Arnav Mishra (7 commits) ## Devpost submission (written by the team) ### Inspiration Most productivity tools tell you what's happening. Plenty of AI tools will summarize your notes or chat with your documents. What I kept running into was a different problem: after a few days of work spread across projects, documents, meeting notes, and tasks, I still had to figure out what actually mattered. The idea behind NexusOS came from that frustration. I wanted something that could understand the state of my workspace, connect information across different sources, and answer a much more useful question: "Given everything that's going on, what should I do next, and why?" That became the foundation for NexusOS. ### What it does NexusOS is an AI operating system for knowledge work. Instead of acting like another chatbot, it observes the current workspace, analyzes projects, tasks, documents, analytics, and relationships inside a knowledge graph, then turns all of that context into an actionable mission. Rather than asking users to trust a black box, every recommendation is backed by evidence. Mission Briefs, Investigation Replay, the Evidence Graph, and the After Action Report all explain what the system found, why it matters, and how it arrived at its recommendation. ### How we built it I built NexusOS as a pnpm monorepo using Next.js, React, TypeScript, Prisma, PostgreSQL, and a shared package architecture. The AI layer is organized as a set of specialized agents that each focus on a different part of the workspace: project analysis, task prioritization, knowledge analysis, risk detection, execution planning, insight generation, and recommendations. Their outputs are combined into a structured work session instead of a single chat response. GPT-5.6 is responsible for the reasoning layer through the OpenAI Responses API, using structured outputs so results can be validated before they're shown in the interface. Codex became part of the development process rather than just a coding assistant. It helped scaffold features, review architecture, refactor components, improve testing, organize the repository, strengthen documentation, and prepare the project for release. Codex was essential for hardening the repository, organizing documentation, improving CI, auditing the workspace, and preparing the project for public release. ### Challenges we ran into The biggest challenge wasn't getting AI to generate recommendations—it was making those recommendations trustworthy. Early versions felt like a smart task manager, but it wasn't obvious why the AI was suggesting something. That pushed me to redesign the experience around evidence instead of answers. Every important recommendation now points back to supporting context through Investigation Replay, the Evidence Graph, and the Mission Brief. Another challenge was turning what started as a hackathon project into something that actually felt like a real product. A surprising amount of time went into improving documentation, testing, repository organization, CI, and the overall developer experience. ### Accomplishments we're proud of I'm most proud that NexusOS doesn't stop at being another AI chat interface. It brings together a multi-agent reasoning pipeline, explainable recommendations, a polished frontend, a typed backend, structured AI outputs, and a repository that someone else can actually clone and run. By the end of the project, it felt much closer to a product than a prototype. ### What we learned One lesson stood out more than anything else: People don't trust AI because it gives answers. They trust it when they can understand where those answers came from. Building features like Investigation Replay and the Evidence Graph reinforced that explainability isn't just a technical feature—it directly changes how confident users feel about acting on AI recommendations. I also gained a much greater appreciation for release engineering. A clean repository, good documentation, reproducible builds, and a polished README are just as important as writing the code itself. ### What's next The current version focuses on demonstrating the core operating-system experience, but there are several directions I'd like to explore next. The biggest step would be integrating with real productivity tools like GitHub, Google Workspace, Slack, Notion, and Linear so NexusOS can reason over live workspace data instead of seeded examples. Beyond that, I'd like to add collaborative mission planning, persistent AI session history, richer knowledge graph exploration, stronger evaluation tooling for agent performance, and production deployment. The long-term goal is to move from an AI assistant that responds to requests to an AI system that continuously understands the state of work and helps people make better decisions before problems become blockers. ## README (from the GitHub repository) # NexusOS **An AI operating system for turning scattered workspace context into executable missions.** NexusOS is a mission-control layer for knowledge work. It analyzes projects, tasks, documents, analytics, knowledge graph signals, risks, and user context, then turns that fragmented workspace state into a focused mission: what to do next, why it matters, what evidence supports it, and how the decision was made. Built for OpenAI Build Week, NexusOS is not a chatbot bolted onto a dashboard. It is a coordinated AI work-session system with specialized agents, structured outputs, an explainable mission brief, investigation replay, an evidence graph, and after-action reporting. ![TypeScript](https://img.shields.io/badge/TypeScript-5.8-3178c6?style=flat-square&logo=typescript&logoColor=white) ![Next.js](https://img.shields.io/badge/Next.js-15-000000?style=flat-square&logo=nextdotjs&logoColor=white) ![React](https://img.shields.io/badge/React-19-61dafb?style=flat-square&logo=react&logoColor=111111) ![Prisma](https://img.shields.io/badge/Prisma-PostgreSQL-2d3748?style=flat-square&logo=prisma&logoColor=white) ![OpenAI](https://img.shields.io/badge/OpenAI-Responses_API-412991?style=flat-square&logo=openai&logoColor=white) ## Preview ![NexusOS mission briefing](assets/screenshots/repository-banner.png) | Dashboard | Mission Command Center | | ------------------------------------------------------ | -------------------------------------------------------------------------------- | | ![NexusOS dashboard](assets/screenshots/dashboard.png) | ![NexusOS mission command center](assets/screenshots/mission-command-center.png) | | Mission Brief | Knowledge Graph | | -------------------------------------------------------------- | ------------------------------------------------------------------ | | ![NexusOS mission brief](assets/screenshots/mission-brief.png) | ![NexusOS knowledge graph](assets/screenshots/knowledge-graph.png) | ## Vision Modern teams do not need another place to store work. They need a system that understands the state of the work, detects what changed, explains what matters, and turns that understanding into the next executable move. NexusOS is built around that idea: AI should not sit beside the product as a chat box. It should operate across the product as a reasoning layer, reading the same projects, tasks, documents, analytics, and knowledge graph that the team uses, then producing decisions that are structured, inspectable, and actionable. The long-term vision is an AI operating system for work: a trusted layer that can plan, explain, prioritize, replay its reasoning, and help teams move from uncertainty to execution without losing the evidence behind each decision. ## Problem Knowledge work is rarely blocked because teams have too little software. It is blocked because the important signals are scattered across too many surfaces: project dashboards, task lists, documents, analytics, settings, notifications, and informal decisions that live in the team's memory. That fragmentation creates three problems: - teams can see activity, but not always the next best action - AI recommendations can feel untrustworthy when the evidence is hidden - dashboards show status, but rarely explain the reasoning behind a change The result is a costly gap between knowing what exists and knowing what to do next. NexusOS is designed to close that gap. ## What Makes It Different NexusOS treats AI as an operating layer for work, not a sidebar assistant. The system loads workspace state, routes it through specialized agents, validates typed outputs, and renders the result as a mission that can be inspected from multiple angles. That makes the core idea memorable: the AI does not just answer a question. It organizes the workspace into a decision, shows the evidence behind that decision, and leaves a reviewable execution record. ## Solution NexusOS turns a workspace into a mission. When a user starts a work session, NexusOS loads the current project, task, milestone, workspace, analytics, insight, settings, and knowledge graph context. It then runs a coordinated set of specialized AI agents that analyze the state of the workspace, identify risks, rank priorities, generate an execution plan, and produce recommendations. The output is not a loose chat response. NexusOS returns a structured mission experience: - a Mission Command Center that shows the AI work session as it runs - a Mission Brief that summarizes the safest next action - an Investigation Replay that explains how the system reached its conclusion - an Evidence Graph that connects projects, tasks, documents, decisions, and knowledge signals - an After Action Report that captures what changed and what should happen next This makes the AI recommendation inspectable. The user can see the action, the reasoning, the evidence, and the expected impact in one continuous workflow. ## Features ### AI Mission System - **Start Work Session** runs the AI operating system pipeline from the product interface. - **Mission Command Center** shows the active AI run, stage progress, elapsed time, completion state, cancellation state, retry state, and result summary. - **Mission Brief** converts the AI result into a clear objective, blocker, confidence score, execution plan, and suggested next action. - **Investigation Replay** lets users step through the reasoning sequence that produced the recommendation. - **Evidence Graph** connects the supporting project, task, document, knowledge, decision, milestone, and timeline signals. - **After Action Report** captures the outcome of the mission and creates a reviewable execution record. ### Workspace Intelligence - **Dashboard** summarizes project health, daily focus, recommendations, notifications, and mission alerts. - **Projects** provides portfolio-level visibility into project status, ownership, risk, progress, tags, and upcoming deadlines. - **Workspace** combines active documents, editable content, task context, and AI conversation history. - **Knowledge Graph** maps workspace relationships and exposes inspectable node details. - **Analytics** shows metrics, trends, distribution, and operational tables with source-aware states. - **Settings** centralizes local AI, notification, appearance, shortcut, and API configuration controls. ### Trust And Explainability - Structured AI outputs validated against typed schemas. - Agent execution trace with prompt IDs, prompt versions, latency, and status. - Repository-backed tools for loading workspace context and persisting generated insights. - Deterministic local provider for reliable development and demo execution. - OpenAI provider adapter for Responses API structured output. ## OpenAI Build Week GPT-5.6 is central to the OpenAI-backed mode of NexusOS. The repository is configured with `OPENAI_MODEL=gpt-5.6` in `.env.example`, and the OpenAI provider adapter uses Responses API structured output so agent results can be validated, persisted, and rendered safely by the product. Codex was essential to the release process: it helped harden the monorepo, organize documentation, audit package manifests, add CI, verify build health, and polish the public GitHub presentation without changing runtime behavior. ## Architecture NexusOS is implemented as a pnpm monorepo with a Next.js application, shared backend packages, a Prisma/PostgreSQL persistence layer, and a dedicated AI orchestration package. ```text NexusOS apps/web Next.js App Router UI API route handlers mission, dashboard, workspace, graph, analytics, and settings screens packages/ai work-session orchestrator specialized agents provider abstraction structured schemas evaluation, memor [README truncated for size] ## Detected evidence (automated analysis) Indexed codebase: 242 recognized source files, 796 KB. - CSS (language) — detected in the code - Next.js (technology) — detected in the code - React (technology) — detected in the code - SQL (language) — detected in the code - Tailwind CSS (technology) — detected in the code - TypeScript (language) — detected in the code - OpenAI (technology) — claimed on Devpost, not found in the code - PostgreSQL (technology) — claimed on Devpost, not found in the code - AI coding agent: Codex — evidence: config files committed to the repository ## Codebase structure (from repository index) ### Files (120 of 276) ``` .agents/README.md .env.development.example .env.example .env.local.example .env.production.example .github/workflows/ci.yml .gitignore .husky/pre-commit .prettierignore .prettierrc.json apps/api/README.md apps/web/app/(app)/analytics/page.tsx apps/web/app/(app)/dashboard/page.tsx apps/web/app/(app)/gallery/page.tsx apps/web/app/(app)/knowledge-graph/page.tsx apps/web/app/(app)/layout.tsx apps/web/app/(app)/loading.tsx apps/web/app/(app)/page.tsx apps/web/app/(app)/projects/page.tsx apps/web/app/(app)/settings/page.tsx apps/web/app/(app)/work-session/page.tsx apps/web/app/(app)/workspace/page.tsx apps/web/app/api/ai/work-session/route.ts apps/web/app/api/analytics/route.ts apps/web/app/api/auth/me/route.ts apps/web/app/api/auth/refresh/route.ts apps/web/app/api/auth/session/route.ts apps/web/app/api/auth/store.ts apps/web/app/api/health/route.ts apps/web/app/api/knowledge/route.ts apps/web/app/api/notifications/route.ts apps/web/app/api/projects/route.ts apps/web/app/api/settings/route.ts apps/web/app/api/tasks/route.ts apps/web/app/api/workspace/route.ts apps/web/app/globals.css apps/web/app/layout.tsx apps/web/components.json apps/web/components/analytics/chart-card.tsx apps/web/components/analytics/donut-chart.tsx apps/web/components/analytics/line-chart.tsx apps/web/components/analytics/metric-widget.tsx apps/web/components/cards/dashboard-card.tsx apps/web/components/cards/insight-card.tsx apps/web/components/cards/metric-card.tsx apps/web/components/cards/project-card.tsx apps/web/components/common/avatar.tsx apps/web/components/common/chip.tsx apps/web/components/common/empty-state.tsx apps/web/components/common/loader.tsx apps/web/components/common/progress-bar.tsx apps/web/components/common/section-header.tsx apps/web/components/common/skeleton.tsx apps/web/components/common/status-badge.tsx apps/web/components/common/tabs.tsx apps/web/components/knowledge/knowledge-graph.tsx apps/web/components/knowledge/node-card.tsx apps/web/components/knowledge/node-panel.tsx apps/web/components/layout/app-shell.tsx apps/web/components/layout/command-palette.tsx apps/web/components/layout/notification-center.tsx apps/web/components/layout/page-layout.tsx apps/web/components/layout/page-placeholder.tsx apps/web/components/layout/sidebar.tsx apps/web/components/layout/top-navigation.tsx apps/web/components/navigation/breadcrumb.tsx apps/web/components/navigation/notification-menu.tsx apps/web/components/navigation/search-bar.tsx apps/web/components/providers/loading-provider.tsx apps/web/components/providers/providers.tsx apps/web/components/providers/theme-provider.tsx apps/web/components/providers/toast-provider.tsx apps/web/components/screens/analytics-experience.tsx apps/web/components/screens/dashboard-experience.tsx apps/web/components/screens/gallery-screen.tsx apps/web/components/screens/knowledge-graph-experience.tsx apps/web/components/screens/projects-experience.tsx apps/web/components/screens/settings-experience.tsx apps/web/components/screens/work-session-experience.tsx apps/web/components/screens/workspace-experience.tsx apps/web/components/settings/select-setting.tsx apps/web/components/settings/settings-section.tsx apps/web/components/settings/toggle-setting.tsx apps/web/components/ui/badge.tsx apps/web/components/ui/button.tsx apps/web/components/ui/card.tsx apps/web/components/ui/input.tsx apps/web/components/ui/select.tsx apps/web/components/ui/separator.tsx apps/web/components/work-session/after-action-report.tsx apps/web/components/work-session/ai-recommendation-card.tsx apps/web/components/work-session/cause-effect-panel.tsx apps/web/components/work-session/demo-scenario-card.tsx apps/web/components/work-session/evidence-graph.tsx apps/web/components/work-session/investigation-replay.tsx apps/web/components/work-session/mission-alert-card.tsx apps/web/components/work-session/mission-brief.tsx apps/web/components/work-session/mission-command-center.tsx apps/web/components/work-session/mission-difference.tsx apps/web/components/work-session/mission-timeline.tsx apps/web/components/work-session/work-session-progress.tsx apps/web/components/work-session/work-session-summary.tsx apps/web/components/work-session/workspace-health-widget.tsx apps/web/components/workspace/ai-chat-panel.tsx apps/web/components/workspace/document-editor.tsx apps/web/components/workspace/document-list.tsx apps/web/hooks/use-mobile.ts apps/web/hooks/use-prefers-reduced-motion.ts apps/web/hooks/use-work-session-runner.ts apps/web/lib/content/gallery.ts apps/web/lib/design-tokens.ts apps/web/lib/mock/analytics.ts apps/web/lib/mock/dashboard.ts apps/web/lib/mock/gallery.ts apps/web/lib/mock/knowledge-graph.ts apps/web/lib/mock/notifications.ts apps/web/lib/mock/projects.ts apps/web/lib/mock/settings.ts apps/web/lib/mock/work-session.ts apps/web/lib/mock/workspace.ts [156 more files omitted for size] ``` ### Dependencies - apps/web/package.json: @nexusos/ai@workspace:*, @nexusos/auth@workspace:*, @nexusos/config@workspace:*, @nexusos/database@workspace:*, @nexusos/prompts@workspace:*, @nexusos/server@workspace:*, @nexusos/shared@workspace:*, @nexusos/types@workspace:*, @types/node@^24.0.0, @types/react@^19.0.0, @types/react-dom@^19.0.0, autoprefixer@^10.4.0, class-variance-authority@^0.7.1, clsx@^2.1.1, date-fns@^4.1.0, eslint-config-next@15.5.20, framer-motion@^12.0.0, lucide-react@^1.24.0, next@^15.0.0, postcss@^8.4.0, react@^19.0.0, react-dom@^19.0.0, react-hook-form@^7.0.0, recharts@^2.15.0, tailwind-merge@^2.5.0, tailwindcss@^3.4.0, typescript@^5.8.0, zod@^3.24.0, zustand@^5.0.0 - package.json: @eslint/eslintrc@^3.3.6, @eslint/js@^9.0.0, @prisma/client@^6.12.0, @types/node@^24.0.0, @typescript-eslint/eslint-plugin@^8.0.0, @typescript-eslint/parser@^8.0.0, dotenv@^16.6.1, eslint@^9.0.0, eslint-config-next@15.5.20, husky@^9.0.0, lint-staged@^15.0.0, prettier@^3.0.0, prisma@^6.12.0, tsconfig-paths@^4.2.0, tsx@^4.19.2, turbo@^2.0.0, typescript@^5.8.0 - packages/ai/package.json: @nexusos/database@workspace:*, @nexusos/prompts@workspace:*, @nexusos/shared@workspace:*, @nexusos/types@workspace:* - packages/auth/package.json: @nexusos/shared@workspace:*, @nexusos/types@workspace:* - packages/config/package.json: @nexusos/shared@workspace:* - packages/database/package.json: @nexusos/auth@workspace:*, @nexusos/shared@workspace:*, @nexusos/types@workspace:*, @prisma/client@^6.12.0, prisma@^6.12.0 - packages/server/package.json: @nexusos/auth@workspace:*, @nexusos/config@workspace:*, @nexusos/database@workspace:*, @nexusos/shared@workspace:* - packages/shared/package.json: @nexusos/types@workspace:* ### Recent commits (newest first) - Polish README for GitHub release - Add repository release assets - Fix typecheck 1.1 - Fix CI typecheck - Fix CI Node version - Fix CI pnpm setup - Release Candidate 1 ## Key source files (fetched from GitHub, selected and truncated for size) ### assets/RELEASE_ASSET_CHECKLIST.md ```markdown # Release Asset Checklist This folder contains public repository assets only. Do not place runtime application assets here unless they are also intended for GitHub, Devpost, or release presentation. ## Ideal GitHub Asset Structure ```text assets/ logos/ nexusos-logo.svg nexusos-logo.png screenshots/ repository-banner.png dashboard.png mission-command-center.png mission-brief.png mission-brief-running.png live-session-work-pipeline.png knowledge-graph.png after-action-report.png demo/ nexusos-demo.gif nexusos-demo.mp4 ``` ## Required Assets | Asset | Target path | Status | Notes | | ---------------------- | ----------------------------------------------- | ------------ | --------------------------------------------------------------- | | Logo | `assets/logos/nexusos-logo.svg` | Not included | No final logo asset was provided. | | Logo PNG | `assets/logos/nexusos-logo.png` | Not included | No final logo asset was provided. | | Repository banner | `assets/screenshots/repository-banner.png` | Included | Derived from the real Dashboard screenshot for the README hero. | | Social preview image | `assets/screenshots/social-preview.png` | Not included | No social preview asset was provided. | | Dashboard screenshot | `assets/screenshots/dashboard.png` | Included | Real Dashboard route screenshot. | | Mission Command Center | `assets/screenshots/mission-command-center.png` | Included | Real active mission command center screenshot. | | Mission Brief | `assets/screenshots/mission-brief.png` | Included | Real mission summary screenshot. | | Knowledge Graph | `assets/screenshots/knowledge-graph.png` | Included | Real knowledge graph screenshot. | | After Action Report | `assets/screenshots/after-action-report.png` | Not included | No after-action report screenshot was provided. | | Animated GIF | `assets/demo/nexusos-demo.gif` | Not included | Excluded for this release. | | Demo video | `assets/demo/nexusos-demo.mp4` | Not included | Excluded for this release. | ## Additional Captures | Asset | Path | Status | Notes | | --------------------- | --------------------------------------------------- | -------- | ---------------------------------------------- | | Mission brief running | `assets/ [truncated — 580 more characters] ``` ### docs/development/changelog.md ```markdown # Changelog Release notes will be tracked here after the initial public release. ``` ### docker-compose.yml ```yaml services: postgres: image: postgres:16-alpine container_name: nexusos-postgres restart: unless-stopped environment: POSTGRES_DB: nexusos POSTGRES_USER: postgres POSTGRES_PASSWORD: postgres ports: - "5432:5432" volumes: - nexusos-postgres-data:/var/lib/postgresql/data healthcheck: test: ["CMD-SHELL", "pg_isready -U postgres -d nexusos"] interval: 5s timeout: 5s retries: 20 volumes: nexusos-postgres-data: ``` ### package.json ``` { "name": "nexusos", "private": true, "version": "0.1.0", "packageManager": "pnpm@11.13.0", "prisma": { "seed": "tsx packages/database/prisma/seed.ts" }, "workspaces": [ "apps/*", "packages/*" ], "scripts": { "dev": "node -e \"require('fs').rmSync('apps/web/.next', { recursive: true, force: true })\" && corepack pnpm --filter @nexusos/web dev", "build": "corepack pnpm --filter @nexusos/web build", "lint": "corepack pnpm --filter @nexusos/web lint", "typecheck": "corepack pnpm --filter @nexusos/web typecheck && tsc -p tsconfig.backend.json --noEmit", "db:generate": "prisma generate --schema packages/database/prisma/schema.prisma", "db:migrate:dev": "prisma migrate dev --schema packages/database/prisma/schema.prisma", "db:migrate:deploy": "prisma migrate deploy --schema packages/database/prisma/schema.prisma", "db:push": "prisma db push --schema packages/database/prisma/schema.prisma", "db:seed": "tsx packages/database/prisma/seed.ts", "db:studio": "prisma studio --schema packages/database/prisma/schema.prisma", "test:integration": "node --require dotenv/config --require tsconfig-paths/lib/register --import tsx --test tests/integration/**/*.test.ts", "test:ai": "node --require dotenv/config --require tsconfig-paths/lib/register --import tsx --test tests/ai/**/*.test.ts", "format": "prettier --write .", "format:check": "prettier --check .", "prepare": "husky", "lint-staged": "lint-staged" }, "dependencies": { "@prisma/client": "^6.12.0" }, "devDependencies": { "@eslint/eslintrc": "^3.3.6", "@eslint/js": "^9.0.0", "@types/node": "^24.0.0", "@typescript-eslint/eslint-plugin": "^8.0.0", "@typescript-eslint/parser": "^8.0.0", "eslint": "^9.0.0", "eslint-config-next": "15.5.20", "husky": "^9.0.0", "lint-staged": "^15.0.0", "dotenv": "^16.6.1", "prettier": "^3.0.0", "prisma": "^6.12.0", "tsconfig-paths": "^4.2.0", "tsx": "^4.19.2", "turbo": "^2.0.0", "typescript": "^5.8.0" } } ``` ### packages/types/package.json ``` { "name": "@nexusos/types", "private": true, "version": "0.1.0", "type": "module", "main": "src/index.ts", "types": "src/index.ts" } ``` ### packages/prompts/package.json ``` { "name": "@nexusos/prompts", "private": true, "version": "0.1.0", "type": "module", "main": "src/index.ts", "types": "src/index.ts" } ``` ### packages/shared/package.json ``` { "name": "@nexusos/shared", "private": true, "version": "0.1.0", "type": "module", "main": "src/index.ts", "types": "src/index.ts", "dependencies": { "@nexusos/types": "workspace:*" } } ``` ### packages/config/package.json ``` { "name": "@nexusos/config", "private": true, "version": "0.1.0", "type": "module", "main": "src/index.ts", "types": "src/index.ts", "dependencies": { "@nexusos/shared": "workspace:*" } } ``` ### packages/auth/package.json ``` { "name": "@nexusos/auth", "private": true, "version": "0.1.0", "type": "module", "main": "src/index.ts", "types": "src/index.ts", "dependencies": { "@nexusos/shared": "workspace:*", "@nexusos/types": "workspace:*" } } ``` ### packages/ai/package.json ``` { "name": "@nexusos/ai", "private": true, "version": "0.1.0", "type": "module", "main": "src/index.ts", "types": "src/index.ts", "dependencies": { "@nexusos/database": "workspace:*", "@nexusos/prompts": "workspace:*", "@nexusos/shared": "workspace:*", "@nexusos/types": "workspace:*" } } ``` [233 more indexed source files omitted to keep this export small. The full file list is in the Codebase structure section above.]