Validate outputs and schemas

Add executable validation at the trust boundaries of OpenTelemetry traces across retrieve → rerank → generate for support answers. Reject unknown fields where they matter, bound string sizes, and coerce only after auth/signature checks when raw bytes are security-relevant. Invariant under test: every span carries trace_id + redacted attrs; prompts sampled ≤ 1%. A TypeScript type or Python annotation is not runtime validation — pair them with parsers.

Automate the fixtures from setup, including a recreation of TRACE-ORPHAN-441. Assert both the visible error and the absence of side effects (no provider call, no queue write, no flag flip). Where metrics matter, assert label enums stay bounded.

Document which failures are client mistakes (4xx) versus operator/config mistakes (5xx/503). Oracle still stands: fixture question produces 3 child spans; total latency attributes sum within 5% of root. Validation should make accidental “success with empty body” impossible for on-call engineer debugging a slow groundedness miss.

Property ideas: shuffled field order, Unicode edges, maximum-length strings, and replayed timestamps. Where money, identity, or citations matter, assertion messages should cite the field name. Do not snapshot entire provider payloads in tests; assert semantically. If validation fails open “to keep the demo working,” you have inverted the lab. Tie at least one CI job to the TRACE-ORPHAN-441 fixture so main cannot regress silently. Re-read every span carries trace_id + redacted attrs; prompts sampled ≤ 1% after each new parser — convenience helpers love to bypass it.

Table-drive status codes and error codes so reviewers see coverage at a glance. Include a Unicode normalization case if user text is accepted. Verify that oversized bodies fail before CPU-heavy work. Where digests or versions are pinned, assert mismatch behavior. Keep golden files small enough to read in review. CI should fail on skipped tests that mark the incident fixture as xfail without a ticket link. In this chapter the product is OpenTelemetry traces across retrieve → rerank → generate for support answers, the human stakeholder is on-call engineer debugging a slow groundedness miss, and the incident id you design against is TRACE-ORPHAN-441. Re-state the oracle in your notes — fixture question produces 3 child spans; total latency attributes sum within 5% of root — and keep the invariant visible: every span carries trace_id + redacted attrs; prompts sampled ≤ 1%. Track trace_completeness ≥ 0.99 and p95_end_to_end_ms ≤ 2500 as the scoreboard. Surface under change control: POST /v1/support/answer. If you only have forty minutes, finish the fixture for missing parent context → orphan spans, cannot join retrieve to answer before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.