···1313The current baseline for saying "the audited suite is green" is:
14141515- `prove -lr t`
1616- - latest full green result in the realigned Meridian worktree: `Files=57, Tests=3051`
1616+ - latest full green result in the realigned Meridian worktree: `Files=58, Tests=3053`
1717- `prove -lv t/server-auth.t`
1818- `perl -c script/differential-validate`
1919- `PERLSKY_RUN_REFERENCE_DIFF=1 prove -lv t/reference-differential.t`
···116116Current suite counts by bucket:
117117118118- `direct reference differential`: `5`
119119-- `audited local regression`: `39`
119119+- `audited local regression`: `40`
120120- `local correctness/infrastructure`: `13`
121121122122| Test file | Bucket | Current note |
···140140| `t/email-confirmation.t` | audited local regression | intentionally testing-friendly email flow plus strict missing-email and invalid-email validation semantics |
141141| `t/email-update-helper.t` | audited local regression | shared email-update helper normalization, token revocation, and duplicate-email error semantics |
142142| `t/event-stream.t` | audited local regression | wire-format, malformed frame, and event decoding coverage |
143143-| `t/extended-api.t` | audited local regression | focused mixed coverage for self-service invite issuance and `applyWrites` after the identity/email, label, and blob/sync happy paths were split out |
143143+| `t/extended-api.t` | audited local regression | focused `applyWrites` happy-path and missing-delete behavior after the self-service invite, identity/email, label, and blob/sync happy paths were split out |
144144| `t/external-handle-update.t` | audited local regression | external-handle update semantics, including DID-resolution checks and empty-body success for external handle adoption |
145145| `t/external-surface.t` | audited local regression | focused external-surface coverage for repo/blob/account-status and missing-blob behavior after splitting discovery and label RPC checks into dedicated suites |
146146| `t/firehose.t` | audited local regression | repo subscription lifecycle, cursor, and CAR behavior |
···169169| `t/repo-api.t` | audited local regression | record mutation and read semantics, but still lighter than ideal on some negative/reference edge cases |
170170| `t/repo-firehose-car.t` | audited local regression | repo commit CAR shape and firehose interactions |
171171| `t/repo_formats.t` | audited local regression | direct repo wire-format and CAR expectations |
172172+| `t/self-service-invite-surfaces.t` | audited local regression | isolated self-service invite issuance coverage for invite minting, account scoping, and account invite-code listing |
172173| `t/sentry.t` | local correctness/infrastructure | Sentry event shaping, stacktrace capture, and dispatcher exception-reporting smoke |
173174| `t/server-auth.t` | direct reference differential | auth/session/service-auth behavior repeatedly compared to official runtime, including bounded create-session password semantics |
174175| `t/service-proxy-local.t` | audited local regression | local appview fallback behavior |
···184185The broadest suites are green and audited, but they still mix several categories of behavior inside the same file:
185186186187- `t/extended-api.t`
187187- Carries real conformance value for `applyWrites`, but it still mixes those with local product behavior such as self-service invite flows.
188188+ Carries real conformance value for `applyWrites`, but it is now much narrower and mostly acts as a focused repo-write regression suite.
188189- `t/external-surface.t`
189190 Carries strong external-surface coverage for repo export, blob access, account-status behavior, and missing-blob listing. It is cleaner after moving discovery and label-RPC checks into dedicated suites, but still remains broader than a single-endpoint conformance file.
190191- `t/uncovered-endpoints.t`