## Summary

Release v1.5.0 — major feature additions, bug fixes, documentation
overhaul, and developer tooling improvements.

### New Features
- **Embeddings API** — `POST /v1/embeddings`, `onEmbedding()`,
`inputText` match field, deterministic fallback embeddings
- **Structured output / JSON mode** — `responseFormat` match field,
`onJsonOutput()` convenience method
- **Sequential responses** — `sequenceIndex` for stateful multi-turn
fixtures, `resetMatchCounts()`
- **Streaming physics** — `StreamingProfile` with `ttft`, `tps`,
`jitter` for realistic timing
- **AWS Bedrock** — `POST /model/{modelId}/invoke` with Anthropic
Messages format translation
- **Azure OpenAI** — provider routing for completions and embeddings
- **Health & models endpoints** — `GET /health`, `GET /ready`, `GET
/v1/models`
- **Docker & Helm** — Dockerfile and Helm chart for Kubernetes
deployment
- **Documentation website** — full docs site at llmock.com
- **Automated drift remediation** — CI-driven drift detection and fix
pipeline
- **`FixtureOpts` / `EmbeddingFixtureOpts`** type aliases exported for
external consumers

### Bug Fixes
- Fix Gemini Live handler crash on malformed `clientContent.turns` and
`toolResponse.functionResponses`
- Add `isClosed` guard before WebSocket finalization events
- Default to non-streaming for Claude Messages and Responses API
(matching real API defaults)
- Fix `streamingProfile` missing from convenience method opts types
- Fix skills/ symlink direction so npm pack includes the write-fixtures
skill
- Fix watcher cleanup on error, empty-reload guard

### Documentation & Tooling
- README rewritten as concise overview with links to docs site
- Write-fixtures skill updated for all v1.5.0 features
- Docs site: Get Started→docs, comparison above reliability, npm version
badge
- ESLint config: add `.worktrees/` to ignores, remove dead CSS

## Test plan
- [x] 813 unit tests pass
- [x] Build succeeds
- [x] Prettier formatting clean
- [x] ESLint clean
- [x] npm pack includes correct files (skills/, .claude-plugin/, dist/,
fixtures/)
- [x] 4 rounds of 7-agent code review — zero actionable findings on
final round

🤖 Generated with [Claude Code](https://claude.com/claude-code)