Measure groundedness and citation validity

Add executable validation at the trust boundaries of citation-first research assistant that maps every claim to evidence IDs. 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: claim without evidence → abstain or drop claim; no orphan sentences. A TypeScript type or Python annotation is not runtime validation — pair them with parsers.

Automate the fixtures from setup, including a recreation of CAP-RESEARCH-FAKE-ID-3. 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: brief on fixture corpus: ≥ 0.95 claims cited; fabricated ID rate 0. Validation should make accidental “success with empty body” impossible for analyst compiling a brief on bike-share policy from a fixed corpus.

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 CAP-RESEARCH-FAKE-ID-3 fixture so main cannot regress silently. Re-read claim without evidence → abstain or drop claim; no orphan sentences 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 citation-first research assistant that maps every claim to evidence IDs, the human stakeholder is analyst compiling a brief on bike-share policy from a fixed corpus, and the incident id you design against is CAP-RESEARCH-FAKE-ID-3. Re-state the oracle in your notes — brief on fixture corpus: ≥ 0.95 claims cited; fabricated ID rate 0 — and keep the invariant visible: claim without evidence → abstain or drop claim; no orphan sentences. Track claim_citation_coverage and fabricated_id_rate as the scoreboard. Surface under change control: POST /v1/research/brief. If you only have forty minutes, finish the fixture for model invents kb://minutes-2099 citation before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.