{"_id":"@aegis-monitor/monorepo","name":"@aegis-monitor/monorepo","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"@aegis-monitor/monorepo","version":"0.1.0","description":"Aegis Monitor - TypeScript/JavaScript SDK for LLM evaluation and governance","author":{"name":"Ezekiel Adetoro"},"keywords":[],"bugs":{"url":"https://github.com/adetorodev/aegis-js/issues"},"homepage":"https://github.com/adetorodev/aegis-js","license":"MIT","private":false,"type":"module","workspaces":["packages/*"],"scripts":{"build":"npm-run-all --parallel build:*","build:core":"npm run build --workspace=@aegis-monitor/core","build:adapters":"npm run build --workspace=@aegis-monitor/adapters","build:scorers":"npm run build --workspace=@aegis-monitor/scorers","build:cost":"npm run build --workspace=@aegis-monitor/cost","build:regression":"npm run build --workspace=@aegis-monitor/regression","build:cli":"npm run build --workspace=@aegis-monitor/cli","test":"npm-run-all test:unit test:integration","test:unit":"npm-run-all test:unit:*","test:unit:core":"npm run test --workspace=@aegis-monitor/core","test:unit:adapters":"npm run test --workspace=@aegis-monitor/adapters","test:unit:scorers":"npm run test --workspace=@aegis-monitor/scorers","test:unit:cost":"npm run test --workspace=@aegis-monitor/cost","test:unit:regression":"npm run test --workspace=@aegis-monitor/regression","test:unit:cli":"npm run test --workspace=@aegis-monitor/cli","test:integration":"vitest run --config tests/vitest.config.ts","test:integration:watch":"vitest watch --config tests/vitest.config.ts","test:integration:coverage":"vitest run --coverage --config tests/vitest.config.ts","docs:api":"typedoc","docs:check":"npm run docs:api","release:verify-packages":"node scripts/release/verify-packages.mjs","release:qa":"npm run lint && npm run format:check && npm run type-check && npm run build && npm run test && npm audit --audit-level=high","release:prepare":"npm run release:qa && npm run release:verify-packages && npm run docs:api","changeset":"changeset","version:packages":"changeset version","lint":"eslint packages --ext .ts,.tsx","format":"prettier --write packages/**/*.ts packages/**/*.json","format:check":"prettier --check packages/**/*.ts packages/**/*.json","type-check":"tsc --noEmit","clean":"npm-run-all --parallel clean:*","clean:core":"npm run clean --workspace=@aegis-monitor/core","clean:adapters":"npm run clean --workspace=@aegis-monitor/adapters","clean:scorers":"npm run clean --workspace=@aegis-monitor/scorers","clean:cost":"npm run clean --workspace=@aegis-monitor/cost","clean:regression":"npm run clean --workspace=@aegis-monitor/regression","clean:cli":"npm run clean --workspace=@aegis-monitor/cli","prepare":"husky install 2>/dev/null || true"},"devDependencies":{"@changesets/cli":"^2.27.9","@typescript-eslint/eslint-plugin":"^7.0.0","@typescript-eslint/parser":"^7.0.0","eslint":"^8.0.0","prettier":"^3.0.0","typedoc":"^0.27.6","typescript":"^5.3.0","vitest":"^1.0.0","@vitest/ui":"^1.0.0","npm-run-all2":"^6.0.0","tsup":"^8.0.0","tsx":"^4.0.0","husky":"^8.0.0"},"engines":{"node":">=18.0.0"},"gitHead":"8c531ead0014050e4f468353a0f8a05e277c2ed2","_id":"@aegis-monitor/monorepo@0.1.0","_nodeVersion":"24.16.0","_npmVersion":"11.13.0","dist":{"integrity":"sha512-VzniOaNAyTKaUO3/OBCHyMsXfP3D/stzNAq89/Jk3Ll+WvTc8GpdTONkekDQm7jmJXDFSdNf7ij+43bwNbwn6w==","shasum":"d3129e551e1412af11b02b4b4283be0c237ee27b","tarball":"https://registry.npmjs.org/@aegis-monitor/monorepo/-/monorepo-0.1.0.tgz","fileCount":195,"unpackedSize":2796831,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIQDU221pP5gdhh/YFVZoyLkC0Ca6j2hmGYWufEcyPtodQQIgM0tGwNaUwmMrUFPU7+D/ffnsY7u8S4w00qk8Sktem5g="}]},"_npmUser":{"name":"adetorodev","email":"adetoroe787@gmail.com"},"directories":{},"maintainers":[{"name":"adetorodev","email":"adetoroe787@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/monorepo_0.1.0_1780930294536_0.7380804968066581"},"_hasShrinkwrap":false}},"time":{"created":"2026-06-08T14:51:34.313Z","0.1.0":"2026-06-08T14:51:34.736Z","modified":"2026-06-08T14:51:34.976Z"},"maintainers":[{"name":"adetorodev","email":"adetoroe787@gmail.com"}],"description":"Aegis Monitor - TypeScript/JavaScript SDK for LLM evaluation and governance","homepage":"https://github.com/adetorodev/aegis-js","keywords":[],"author":{"name":"Ezekiel Adetoro"},"bugs":{"url":"https://github.com/adetorodev/aegis-js/issues"},"license":"MIT","readme":"# Aegis Monitor – JavaScript/TypeScript SDK\n\n> A TypeScript-first, provider-agnostic evaluation and governance library for Large Language Model (LLM) applications.\n\n## Overview\n\nAegis Monitor enables JavaScript and TypeScript applications to:\n\n- **Evaluate** prompts against datasets\n- **Track** token usage and cost\n- **Detect** quality regressions\n- **Compare** models side-by-side\n- **Enforce** evaluation thresholds in CI pipelines\n\nWith major provider support including **OpenAI**, **Anthropic**, and **Google**.\n\n## Key Features\n\n✨ **TypeScript-First** – 100% typed public API\n🚀 **Provider-Agnostic** – Works with any LLM provider\n📊 **Regression Detection** – Catch quality drops before production\n💰 **Cost Tracking** – Understand the cost per evaluation\n🔄 **CI/CD Integration** – Enforce thresholds in your pipeline\n⚡ **Performance** – <50ms overhead per request\n🎯 **Edge Compatible** – Runs on Node.js, Vercel, Cloudflare, etc.\n🔒 **No Telemetry** – Complete privacy, runs 100% locally\n\n## What It Is Not\n\nThis is a **pure SDK** – no dashboard, no hosted service, no telemetry.\n\n## Installation\n\n```bash\nnpm install @aegis-monitor/core @aegis-monitor/adapters @aegis-monitor/scorers\n```\n\n[Full installation guide](./docs/INSTALLATION.md)\n\n## Quick Start\n\n```typescript\nimport { Evaluator } from '@aegis-monitor/core';\nimport { OpenAIAdapter } from '@aegis-monitor/adapters';\nimport { ExactMatchScorer } from '@aegis-monitor/scorers';\n\nconst dataset = {\n  cases: [\n    { input: 'What is 2+2?', expectedOutput: '4' },\n  ],\n};\n\nconst adapter = new OpenAIAdapter({\n  apiKey: process.env.OPENAI_API_KEY,\n  model: 'gpt-4',\n});\n\nconst evaluator = new Evaluator(adapter, [new ExactMatchScorer()]);\nconst result = await evaluator.evaluate(dataset);\n\nconsole.log(result.metrics);\n```\n\n[Full quickstart guide](./docs/QUICKSTART.md)\n\n## Documentation\n\n- [Installation Guide](./docs/INSTALLATION.md)\n- [Quick Start](./docs/QUICKSTART.md)\n- [API Reference](./docs/api/)\n- [Provider Setup](./docs/guides/PROVIDER_SETUP.md)\n- [Custom Scorers](./docs/guides/CUSTOM_SCORER.md)\n- [Custom Adapters](./docs/guides/CUSTOM_ADAPTER.md)\n- [CI/CD Integration](./docs/guides/CI_CD.md)\n- [Examples](./examples/)\n\n## Architecture\n\n```\nAegis Monitor SDK\n├── Adapter Layer (OpenAI, Anthropic, Google)\n├── Evaluation Engine\n├── Scoring Engine\n├── Cost Engine\n├── Regression Engine\n└── CLI Wrapper\n```\n\n## CLI\n\n```bash\nnpx aegis-monitor init\nnpx aegis-monitor run --config aegis.config.ts\nnpx aegis-monitor baseline save --config aegis.config.ts\nnpx aegis-monitor compare --config aegis.config.ts\n```\n\n## Development\n\n```bash\n# Clone repository\ngit clone <repo>\ncd aegis-js\n\n# Install and build\nnpm install\nnpm run build\n\n# Run tests\nnpm run test\n\n# Code quality\nnpm run lint\nnpm run format\nnpm run type-check\n```\n\nSee [DEVELOPMENT_PLAN.md](./DEVELOPMENT_PLAN.md) for the implementation roadmap.\n\n## Core Philosophy\n\n> LLMs must be treated like production dependencies.\n>\n> If you wouldn't deploy untested code, you shouldn't deploy untested prompts.\n\n## License\n\nMIT\n\n## Contributing\n\nSee [CONTRIBUTING.md](./CONTRIBUTING.md) for guidelines.\n\n","readmeFilename":"README.md","_rev":"1-4485cfbc1ea6f619b8cf20d0b4b9d809"}