You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The capability-vulnerability audit (docs/audits/capability-vulnerability-audit.md) states, for each record, the trust boundary it crosses, the control whose absence makes it exploitable, and a three-line capability, vulnerability and impact summary. Only AVE-2026-00081 to 00089 carry them as data (security_boundary, missing_control, vulnerability_rationale); the quality queue (docs/decisions/quality-queue.json, #410) puts the other 80 in batches of ten.
This is batch 7: 00077, 00078, 00079, 00080 (the 4 records of the queue's batch 7 that need no decision; the rest of that batch waits for the maintainer's decisions).
Proposal, one pull request
Add the three fields to each of the 4 records, additively, in the shape AVE-2026-00087 uses, with the content taken from the audit's own sections for that record and tightened to the template's style. The audit's commentary (its cross-references to other records and its decision text) stays in the audit. last_updated moves, as CONTRIBUTING asks for an improved record.
Out of scope, because they are content changes that need the maintainer's approval or belong to other work: fingerprint rewording (clarify_fingerprint, reframe_fingerprint, rewrite), scores, framework mappings, and primary sources. No credit correction is included: the researcher-credit audit is not in the repository.
Activity
fix: quality batch 7, taxonomy fields for 4 records (#438) (#439)