Thin Palantir Foundry adapters and model catalog for the Vercel AI SDK.
Use Node.js 24 and the pnpm version declared in package.json.
pnpm install --frozen-lockfile
pnpm run verifyverify runs release-policy checks, Biome, Vitest, typecheck, build, skill validation, and package-content audit through Nx. CI also tests the packed library with AI SDK 6, SDK 7 stable, and the supported late SDK 7 beta line.
| Test layer | Coverage | CI |
|---|---|---|
| Unit | Config, catalog, adapter behavior, harness tooling | Yes |
| Integration | Offline SDK tool loop, telemetry, packed-library compatibility | Yes |
| API end-to-end | Manual Foundry capability harness | No; requires credentials |
| Web end-to-end | None | No |
Run pnpm run test:live -- --model openai:gpt-5-nano for a focused live probe. Use --catalog to survey the catalog. The test status page explains evidence limits and local artifacts.
Provider factories stay thin: config and catalog exports live at the package root, provider clients use separate subpaths, and applications compose registries with AI SDK createProviderRegistry. Base examples are shared with the published agent skill through the examples/base symlink.
Copyright 2026 SHPIT LLC. Apache-2.0. See security reporting.