Topic · defect review current

Defect review current

Gist

Defect review current is whether an active spec with implemented_by still has a live defect_review stamp: no stamp, stale fingerprints, claims without outcomes, verified claims never promoted. Proof runs proof audit --check defect_review_current. Jama still authors.

proof audit --check defect_review_current

Keep the Salesforce status page if that is the search. Keep Jama if it already authors the shall. Neither one asks whether this spec was claim-reviewed after the code moved.

01 · The silent no-stamp

A spec with code links and no defect_review block still looks reviewed if no KnownIssue was filed yet.

Presence of the shall is not this hop. This hop is whether the claim-review lifecycle still has a machine-checkable stamp.

The check is defect_review_current. Stage verify. Default severity is warning, and it is on unless you opt out. It walks every active spec with implemented_by links inside the effective verification scope. A spec with no code links is out of scope. The requirement stamp is the primary state. The audit never dispatches a review and never spends metering budget.

Every adjacent hop in Proof is scoped to a different object. Known issues reviewed asks whether an open KnownIssue still has a current review_date. It is quiet when the directory is empty. Known issue complete asks whether a filed KI has evidence and an origin. It never asks whether a claim was reviewed on the spec. Documented claim verified asks whether a documented behavior still has a passing witness. It is not the defect-claim lifecycle. Leave SYS-REQ-904 with implemented_by, never run proof claims plan, and those hops stay green. That is the hole.

id: SYS-REQ-904
status: active
implemented_by:
  - pkg/claims/check.go
# no defect_review block

Six faces fire, in lifecycle order. No stamp: stage, review, then proof claims finalize. Stale fingerprint: the current tree disagrees with the stamp (spec YAML excluding the stamp, a bundled file, claims input, prompt, or the stamp digest itself). Unverified claims: an emitted claim has no outcome. Unpromoted claims: a verified claim was never linked as source_claim on a KnownIssue. Undismissed claims: a dismissal has no stated reason. Partial session: a cell in the session errored. Clean is counted as N/M specs: claim-reviewed.

A project with no stamps, no claim objects, no sessions under proof/claims/sessions/, and no adoption flag sees a clean no-op: no defect-review sessions (not adopted). Declaring adoption before the first session lands is a YAML flag, not a review.

proof claims plan --spec SYS-REQ-904
proof claims run --session <id>
proof claims finalize --session <id>
proof audit --check defect_review_current

Route B is legitimate: write the same cell artifacts yourself, then finalize with --via agent-review. The stamp records the route so the two paths stay comparable. Deleting the spec YAML also silences the hop. It destroys the finding rather than recording the decision.

02 · The exhibit

Same SYS-REQ-904. Silent no-stamp, or this hop.

Active spec, code links present, no defect_review stamp, no KnownIssue filed. Click the tabs.

The row

  • Ask does SYS-REQ-904 exist, sit active, and name implemented_by
  • Stamp the YAML is present, no defect_review block, proof/known-issues is empty
  • Why known-issues-reviewed is a pass on zero files
Status green

This hop

Nobody asked whether the spec was claim-reviewed after the code landed. The finding kind is this hop.

No stamp

The row

Keep the Salesforce bulletin. Keep the Jama field. That is not this hop.

Keep the record

Proof

  • Ask does SYS-REQ-904 still carry a live defect_review stamp
  • Out SYS-REQ-904 defect review: no requirement stamp. stage: proof claims plan --spec SYS-REQ-904
Stamp unread

Same SYS-REQ-904. Silent no-stamp, or this hop. Click the tabs.

Surface What they do What Proof does What we lose
Known issues reviewed Warning. Whether an open KnownIssue still has a current review date. Zero files is a pass. Whether the spec was claim-reviewed before a KI existed. Not the review_date hop. See known issues reviewed.
Known issue complete Whether a filed KI has evidence and an origin. Whether a verified claim was promoted onto that surface at all. Not the completeness hop. See known issue complete.
Documented claim verified Whether a documented behavior still has a passing witness. Whether comments and docstrings near the function were reviewed as claims. Not the documented-claim hop. See documented claim verified.
Salesforce known issue Ads known issue 1600 is that status page if it ranks. A stamp on this spec, next to this shall. Not Salesforce. We have not frozen a vendor bulletin pack.
Jama field The authoring programme. Attributes if you put them there. A YAML stamp the audit can name next to the shall. Not Jama's V&V. Jama still authors. We have not run a frozen Jama pack.

The teaching graph is still one active spec next to code and no stamp. Stage it, review it, then record the stamp.

proof claims plan --spec SYS-REQ-904
proof claims finalize --session <id>

The other honest exits are outcomes, not silence:

proof claims disposition --claim <session>/<cand-id> --outcome verified --evidence <path>
proof known-issue new ...  # then source_claim: <session>/<cand-id>
proof claims disposition --claim <session>/<cand-id> --outcome dismissed --reason "setup comment, not a defect"

KnownIssue cadence stays on known issues reviewed. Completeness stays on known issue complete. Documented behavior stays on documented claim verified. Do not treat a Salesforce bulletin as this cell. Jama still authors. Proof vs Jama.

03 · The honest loss

Proof names a missing stamp. It does not review the claim, and it does not prove the Go.

A quiet proof audit --check defect_review_current can still mean the project was never adopted. Default is warning, not fail. Jama still authors.

The hop does not dispatch Route A, does not write a ledger, and does not spend metering. A light content floor on verified evidence is not both-ways verification; that discipline stays with the author. --outcome deferred satisfies the adjudication face and is re-surfaced in the summary counts; it is not terminal. Schema-invalid emit lines and upstream ticket citations ride as disclosures, marked [info], never as a warning face. Recording a review never invalidates the stamp: the spec fingerprint is computed with the defect_review block removed. Claims themselves are ephemeral under .proof/defect-sessions/. Zero in-scope specs, or a project that never adopted, is a pass that inspected nothing of this kind. A finding is not a waiver: proof waive is a human authorization gate. The hop does not prove the Go. We have not scored this floor against a frozen Jama pack or a Salesforce known-issue export. The loss is named, not scored.

The KnownIssue cadence stays on known issues reviewed. The documented-claim hop stays on documented claim verified. The engagement stays on software correctness audit. Jama still authors.

04 · Nearby questions

What people type next.

What is defect review current? Same question. Same URL.

Is this known issues reviewed? No. That hop is whether an open KnownIssue still has a current review date. This hop is whether the spec was claim-reviewed before a KI existed. See known issues reviewed.

Is this known issue complete? No. That hop is whether a filed KI has evidence and an origin. This hop is whether a verified claim was promoted onto that surface. See known issue complete.

Is this documented claim verified? No. That hop is whether a documented behavior still has a passing witness. This hop reviews comments and docstrings as claims. See documented claim verified.

Is this a Salesforce known issue? No. Ads known issue is that status page. Ads defect review is 10 and generic. A hosted vendor bulletin is not this cell.

Does an unadopted project fail? No. No stamps, no sessions, no adoption flag is a no-op pass.

Does deferred fail? No. Deferred satisfies adjudication and stays visible in the summary. Promote or dismiss later.

Does a green hop prove the Go? No. The hop does not run the review backend. It does not prove the function.

Is Proof a Jama alternative for the shall? No. Jama still authors. Proof vs Jama.