Repository navigation
fix(OMN-17184): close all 14 omnimemory Dependabot alerts (8 high) - #467
Merged
Merged
Conversation
Lockfile-only remediation of every open Dependabot alert on omnimemory (8 high / 4 medium / 2 low). No source or pyproject change — every bump sits inside the constraints already declared in pyproject.toml. - Mako 1.3.10 -> 1.4.1 CVE-2026-41205, CVE-2026-44307 (2x high) - pyasn1 0.6.3 -> 0.6.4 CVE-2026-59884/59885/59886 (3x high) - urllib3 2.6.3 -> 2.7.0 CVE-2026-44431, CVE-2026-44432 (2x high) - deepdiff 8.6.1 -> 9.1.0 CVE-2026-33155 (high) - mcp 1.27.2 -> 1.29.1 CVE-2026-59950 (high, direct) - tuf 6.0.0 -> 7.0.0 GHSA-qp9x-wp8f-qgjj (medium) - requests 2.32.5 -> 2.34.2 CVE-2026-25645 (medium) - h2 4.3.0 -> 4.4.1 CVE-2026-71554 (medium) - idna 3.11 -> 3.19 CVE-2026-45409 (medium) - Pygments 2.19.2 -> 2.21.0 CVE-2026-4539 (low) Two majors are deliberate and both are transitive-only: - deepdiff 8 -> 9 matches the version omnibase_infra already runs on dev, and omnimemory has no direct deepdiff import. - tuf 6 -> 7 is pulled through sigstore; nothing here imports tuf directly. mcp was deliberately held INSIDE the 1.x line. pyproject declares `mcp>=1.27.2,<3.0.0`, so an unconstrained upgrade resolves to 2.1.1 and drags in httpx2, httpcore2, httpx2-jsfetch and mcp-types — a transport swap, not a security patch. Alert #47 only needs >= 1.28.1, so the bump was pinned to `<2` and lands on 1.29.1 with no new packages. Verification: `uv lock --check` clean; `uv sync --all-extras` resolves; tests/unit run in full — 1233 passed, 0 failed. Ticket: OMN-17184
|
Important Review skippedReview was skipped due to path filters ⛔ Files ignored due to path filters (1)
CodeRabbit blocks several paths by default. You can override this behavior by explicitly including those paths in the path filters. For example, including ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Comment |
jonahgabriel
pushed a commit
to OmniNode-ai/onex_change_control
that referenced
this pull request
Aug 31, 2026
…imemory#467 (#7794) * evidence(OMN-17184): author OCC companion for OmniNode-ai/omnimemory#467 OCC companion by node_pr_lifecycle_fix_effect (OMN-13317 F1 / OMN-13990 / OMN-14285). Product PR head b85725aba58e5fab6fe8687566a23b7fa7b2f23f. * evidence(OMN-17184): self-bind OCC#7794 + rebind contract_sha256 --------- Co-authored-by: omnimarket-bot <bot@omninode.ai>
jonahgabriel
added a commit
that referenced
this pull request
Aug 31, 2026
…pattern fixes (#470) omnimemory#466 (OMN-17236) fixed three PII detector pattern defects on a live persistence path and merged to dev at 2026-08-30T17:50:52Z. The v0.18.0 tag and PyPI publish had already happened at 14:06Z that day, so the fixes missed the cut by under four hours and there has been no release since. The consequence is live and reproducible in the .201 dev lane, whose containers all report omninode-memory 0.18.0: docker exec omninode-runtime python -c "... PIIDetector().detect_pii(s) ..." visa -> 'card ***-***-****111111 end' types: ['phone'] discover -> 'card ***-***-****111117 end' types: ['phone'] uuid -> 'run 550e8400-e29b-41d4-a***-***-****40000 done' ssn-sp -> 'ssn 123 45 6789 end' types: [] The trailing six digits of a Visa or Discover card are still being written into stored content, UUID-bearing storage requests are still refused, and a space-separated SSN is still not detected at all. This bump is the first of three links; the omnimarket repin (exact pin at ==0.18.0) and the dev-lane image rebuild follow on OMN-17390. Release contents (v0.18.0..dev): #466 PII fixes + 170 behavioral tests, #467 all 14 Dependabot alerts, #464 omnibase-infra 0.38.14, #468 sibling locks, #469 CI guide doc fix.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Closes all 14 open Dependabot alerts on
omnimemory(8 high / 4 medium / 2 low). Lockfile only — no source orpyproject.tomlchange; every bump sits inside constraints already declared.Ticket: OMN-17184
Makopyasn1urllib3deepdiffmcprequeststufh2idnaPygmentsmcpwas deliberately held inside 1.xpyproject.tomldeclaresmcp>=1.27.2,<3.0.0, so an unconstrained upgrade resolves to 2.1.1 and drags inhttpx2,httpcore2,httpx2-jsfetchandmcp-types— a transport swap, not a security patch. Alert #47 only needs>= 1.28.1, so the bump was pinned<2and lands on 1.29.1 with zero new packages.Two majors are deliberate and both are transitive-only:
deepdiff8→9 matches the version omnibase_infra already runs on dev, andtuf6→7 arrives throughsigstore— nothing here imports either directly.Verification
uv lock --checkclean;uv sync --all-extrasresolvesuv run pytest tests/unit— 1233 passed, 0 failedPREPUSH_*override, no--no-verify.Context
One of six PRs closing 141 open Dependabot alerts org-wide under OMN-17184. The ticket originally scoped only omnidash's 39; a live sweep of all 14 registry repos found 141 across six.
Evidence-Ticket: OMN-17184
Evidence-Source: OCC#7794