Proof Portal

Projects

jsonparser

ProbeLabs36 findings · 123 requirements
Latest run · Audit @ 2026-08-03
Audited Aug 3, 2026commit 775a6cc4f01 errors · 9 warnings
Verification
Needs attention1 error9 warnings
116/116 realizable123/123 meet policyverification complete: 123

as of 775a6cc · Aug 3, 2026 · proof-demo

Audit progress
How far along this run's proof processing is

Formalization

Formalized116 / 12394%
7 not yet formalized
Vacuity checked123 / 123100%
checked ok

Realizability

Realizable116 / 116100%
of 116 evaluated
Unrealizable0 / 1160%
specs that cannot be satisfied — action needed
Realizability undecided0 / 1160%

Verification

Verification complete123 / 123100%

Assurance level

Requirements by verification rigor (A highest → E lowest, inspired by DO-178C). A distribution of requirements — not a count of checks; the rows sum to the total requirement count.

Level B — High rigor20 / 12316%
Level E — Minimal rigor103 / 12384%

Traceability

Meets verification policy123 / 123100%
every requirement has the evidence its policy requires
Required gaps0
no required trace link is missing

Audit checks

Checks passing148 / 16789%
4 skipped
Errors1
must resolve
Warnings9
Advisory2
informational, non-gating
View all checks
Open findings by severity
Open findings needing attention in this run · 13 resolved not shown
Check highlightsAll 167 checks
Checks reporting issues or not passing in this run
Mediumauthored_delta_expected

2 traced production implementation files lacked current no-authored-change review

Mediumchange_evidence_complete

2 in-branch-changed change-typed reviews lack the backing evidence their declared change kind requires (refactor: no-authored-change; fix: a resolvable DEFECT; feature: a feature LocalChange + spec update)

Mediumcode_mcdc_coverage

code-level MC/DC below policy (aggregate decisions 0.0% < 100%, aggregate conditions 0.0% < 100%, parser decisions 0.0% < 100%, parser conditions 0.0% < 100%)

Highhigh_severity_reproducer_grade

6 of 6 high/critical-severity or CVE-surface known issue(s) fail the reproducer-grade gate (3 missing/wrong severity_basis; 0 missing evidence_manifests; 0 non-resolving evidence_manifests; 0 missing tripwire_mutation; 6 reproducer does not exercise affected_api; 9 total gap(s)) — a high-stakes claim shall NOT rest on static analysis or judgment alone: set severity_basis=reproducer, declare a tripwire_mutation, and attach a resolving evidence_manifest whose reproducer exercises the target (issues #350, #845)

Mediumknown_issue_complete

10 of 10 active known issue(s) below the quality floor (0 missing evidence; 0 missing origin; 0 unresolvable origin; 0 inconsistent defect_class; 0 unclassified defect_class; 0 missing `// Reproduces:` comment; 0 missing/invalid CVSS; 0 severity_basis gap(s); 0 environment_note-only at high assurance; 30 stale/invalid evidence; 0 missing current reproducer evidence; 30 total gap(s); 0 dedup match(es) against 0 current KI(s) (ACC-08, integrated)) — a KI without evidence, a recorded origin, the in-test link comment, a valid CVSS vector (security-relevant), an attributed severity basis (issue #277), or — at high project assurance — anything beyond a prose `environment_note` (issue #322) is a sticky note, not a tracked defect; stale evidence hashes and missing current reproducer-evidence rows (issue #392) now fail the audit the same way `proof known-issue check --fail` does; possible duplicates against prior-audit / sibling KIs (ACC-08) are now surfaced here as the same tracked-debt shape

Mediumknown_issue_template_transfer

5 of 10 open known issue(s) need template-transfer hygiene (0 incomplete transfer; 0 suggest template_class; 5 missing dedup_armor)