Skip to content

docs: fix-verification pass over the review-remediation batch (#50–#65) - #66

Merged
tyler-rich merged 1 commit into
devfrom
claude/review-fix-verification-k7k2tr
Jul 13, 2026
Merged

docs: fix-verification pass over the review-remediation batch (#50–#65)#66
tyler-rich merged 1 commit into
devfrom
claude/review-fix-verification-k7k2tr

Conversation

@tyler-rich

@tyler-rich tyler-rich commented Jul 13, 2026

Copy link
Copy Markdown
Owner

Summary

Read-only verification of the review-remediation batch (PRs #50#65) against the current merged state of dev (e545d77), not against what any report claims. Adds docs/reviews/fix-verification.md. Docs only — no application code touched.

Headline results

  • HIGH: 10/11 RESOLVED with a covering test. One genuine miss: H5 / CON-4 — the scanner JSON parse still runs on the event loop (scanners/trivy.py:285, grype.py:186, base.py:362), and it has no docs/ARCHIVE.md §14 entry at all (fell through the gap between the "Top 5 Phase 1 — Auth & secrets foundation #2" and "CON-5–CON-20" change-sets). Not fixed, not logged as deferred.
  • MEDIUM: 26/26 addressed — all RESOLVED. Caveat: M19/P1-2 is RESOLVED but untested (no panel-level test); M22/M23 image bumps couldn't be boot-verified in the egress-restricted env.
  • LOW: 24/25 resolved. Remaining: L24/SC-11persist-credentials: false missing on ci.yml (partial, low risk).

Decision-specific checks (all landed as decided)

  • M2/SEC-3 — entropy-floor validation only; encrypt/decrypt paths, HKDF, and token format untouched; regression test proves existing ciphertext still decrypts. No v2 KDF/re-encryption/format change.
  • M11/CON-10 — BOTH fixes present (pool sizing + cap le=32, AND the connection-release-after-atomic-claim refactor).
  • M23/SC-7 — tecnativa bumped to v0.4.2; wollomatic migration NOT done in-batch, tracked in issue Evaluate migrating the Docker socket proxy from tecnativa to wollomatic/socket-proxy #63.
  • H11/SC-1 — uv build/dev-time only (not a runtime dep); plain-pip --require-hashes install; CI drift check present.
  • CON-11/M12 — retired once by the watchdog; verified, not double-applied.

Regressions / conflicts checked — all clean

CON-10 session threading in _run/_dispatch/_persist/_fail/_notify; CSP vs. the SPA; the two "Run now" fixes (L8/CON-17 ↔ L12/APIR-7, no double-apply); restore/backup pair (M10/CON-9 ↔ CON-3/SEC-2, no conflict). CLAUDE.md ↔ §14 drift (R1–R8, D1, D2) confirmed closed.

Dependabot triage (2 alerts: 1 high, 1 moderate on default branch)

Default branch is main @ v0.1.0; the entire remediation batch (H11 lockfile, SC dependency bumps, #51 dogfood CVE fixes) lives on dev and is unpromoted. The alerts are category (a) — already fixed on dev, pending a release promotion — not genuinely new CVEs. Recommended: promote devmain. (The exact advisory IDs couldn't be enumerated from this read-only session — the Dependabot alerts endpoint isn't exposed by the available tooling; the triage is structural.)

See docs/reviews/fix-verification.md for the full regressions/needs-attention section, HIGH/MEDIUM tables, LOW backlog, and decision-verification detail.

Not merging — for your review.

Read-only verification of PRs #50-#65 against the current merged state of dev.
Confirms 10/11 HIGH, 26/26 MEDIUM (as addressed), and 24/25 LOW findings resolved
with covering tests, and that the five mid-batch decisions (M2, M11, M23, H11,
CON-11) landed as decided. Flags the one genuine miss (H5/CON-4, scanner JSON
parse still on the event loop, with no deviation entry), triages the two
Dependabot default-branch alerts as already-fixed-on-dev-pending-promotion, and
notes the L24/SC-11 and M19-untested residuals.
@tyler-rich
tyler-rich merged commit 8e100ee into dev Jul 13, 2026
4 checks passed
@tyler-rich
tyler-rich deleted the claude/review-fix-verification-k7k2tr branch July 31, 2026 04:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant