chore(root): Add LGTM observability suite

[rifont]

Note

Medium Risk
Touches core chat/WebSocket execution paths (auth middleware ordering, room join semantics, RPC request/response handling) and introduces new external compaction calls + telemetry metrics, which could affect reliability/latency if misconfigured. Changes are mostly additive with expanded test coverage, lowering regression risk but still warranting careful review in production.

Overview
Observability: Adds a new TelemetryIngestModule with POST /v1/telemetry/ingest to accept batched client metrics, wires HealthModule into the API, and updates the API Docker image to auto-bootstrap OTEL and expose the Prometheus port (9464). CI is extended with knip dead-code analysis and a PR-only bundle size report, with .size-limit.json ignored.

Chat execution hardening + metrics: Refactors chat RPC to use Socket.IO emitWithAck (removing the server-side pendingRequests/rpc_response event flow), enforces per-chatId ownership by userId, and adds connection/RPC message-size/duration metrics + SSE event/connection counting. Chat streaming adds TTFT callbacks, newline/LaTeX normalization transforms, higher recursionLimit, and supports a contextPayload passed into the agent via a new createClientContextMiddleware (skills/memory injected without extra RPC).

Geometry testing/analysis: Moves GLB analysis and requirement evaluation logic out of the Nest service into @taucad/testing/geometry, and expands measurement tests with a new connectedComponents check coverage.

^{Reviewed by Cursor Bugbot for commit 796e401. Bugbot is set up for automated code reviews on this repo. Configure here.}

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: adce4b99-90b5-4900-b6cf-0c60eb36b4a9

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch observability-v1

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[rifont]

@cursoragent review

[github-actions]

API Preview 🐙

Status: ❌ Deployment Failed
URL: https://tau-api-pr-169.fly.dev

[github-actions]

UI Preview 🎈

Status: ❌ Deployment Failed
URL: https://tau-ui-pr-169.fly.dev/

[rifont]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[nx-cloud]

View your CI Pipeline Execution ↗ for commit b5b25ae

Command	Status	Duration	Result
nx affected -t lint test build typecheck size	❌ Failed	7m 36s	View ↗
nx run scripts:validate-frontmatter	❌ Failed	<1s	View ↗
nx build runtime	✅ Succeeded	1s	View ↗
nx run scripts:validate-project-names	✅ Succeeded	<1s	View ↗

---

☁️ Nx Cloud last updated this comment at 2026-03-24 10:36:45 UTC