Skip to content

fix(release): use Node patch that rejects compromised CA keys - #69

Merged
joshavant merged 2 commits into
mainfrom
security/emudhra-node-runtime
Oct 22, 2026
Merged

joshavant merged 2 commits into
mainfrom
security/emudhra-node-runtime

Conversation

@joshavant

Copy link
Copy Markdown
Contributor

What Problem This Solves

Release jobs select Node24.21.0, which can still accept the compromised eMudhra keys.

User Impact

Release jobs use the fixed Node24.22.1 patch. Hosted CI remains required. Linux arm64 incident qualification does not establish native macOS/Windows runtime coverage. Installed consumer runtimes and provider-owned action/runner runtimes are separate obligations. The intentional Node25 CI compatibility lane remains an upstream-release wait; this PR preserves its major.

Why This Change Was Made

Update the release workflow’s existing NODE_VERSION input to the independently qualified fixed24.22.1 patch. Package runtime compatibility and public APIs are unchanged.

Evidence

pnpm check passes, including the release package content check. Workflow YAML and actionlint pass; git diff --check and independent P2 review pass. The fixed Node binaries independently reject all12 incident chains in8 trust modes.

@joshavant
joshavant requested a review from a team as a code owner October 21, 2026 22:09
@clawsweeper

clawsweeper Bot commented Oct 21, 2026 •

Copy link
Copy Markdown

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

@clawsweeper clawsweeper Bot added clawsweeper:intake:bug Preliminary submission classification; full source review pending. clawsweeper:intake:needs-repro Preliminary submission classification; full source review pending. P2 Normal priority bug or improvement with limited blast radius. merge-risk: 🚨 security-boundary 🚨 Merging this PR could weaken sandboxing, authorization, credentials, or sensitive data. rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. and removed clawsweeper:intake:bug Preliminary submission classification; full source review pending. clawsweeper:intake:needs-repro Preliminary submission classification; full source review pending. labels Oct 21, 2026
@clawsweeper

clawsweeper Bot commented Oct 21, 2026

Copy link
Copy Markdown

Codex review: needs real behavior proof before merge.

What this changes

This PR pins release tooling to Node 24.22.1 and documents its rejection of compromised eMudhra certificate keys.

Example: The release job sets up Node before installing dependencies and accessing npm.

  • Before: The version selector is 24, allowing an older cached Node 24 patch.
  • After: The version selector requires 24.22.1, which includes upstream certificate-key blocking.

Review scores

Measure Result What it means
Overall readiness 🧂 unranked krab (1/6) The patch is minimal and source-supported, with passing checks and no concrete defect; required security behavior evidence is absent.
Proof confidence 🧂 unranked krab (1/6) Real behavior proof is necessary before merge. See Before merge.
Patch quality 🦐 gold shrimp (3/6) Security review found an item that needs attention.

Product

Kind: Bug fix · Worth it: Yes · Fix scope: Complete
User problem: Release tooling can make HTTPS requests using a Node patch that still trusts compromised certificate keys.
Reason: The supplied review instructions establish maintainer-approved direction, and the narrow runtime pin addresses a documented upstream security defect without changing consumer runtime support.

Merge readiness

⛔ Blocked before merge - 3 items remain

Keep this PR open: it makes a focused security improvement that is absent from main, and inspection found no introduced code defect.

Priority: P2
Reviewed head: cd6172b21a13b80f0b37baebe86fccb6cb715bad

Before merge

  • Add real behavior proof - Authority-chain proof required: the body claims Node 24.22.1 rejects twelve incident chains in eight trust modes, but includes no inspectable output or artifact connecting the selected release runtime to successful authorized requests and compromised-key rejection before authenticated HTTP effects; attach redacted terminal output or logs, excluding credentials and non-public endpoints. After adding proof, update the PR body, then request a new review with @clawsweeper re-review. The PR author or someone with repository write access can request it.
  • Resolve security review attention item - The patch adopts upstream TLS key revocation without expanding workflow permissions, but its authority-boundary evidence remains incomplete.
  • Resolve merge risk - Before merge, establish that the selected runtime rejects a compromised-key peer before authenticated registry HTTP requests while permitting a valid peer.

Findings

None.

Tests

  • Missing end-to-end proof: Attach the claimed incident qualification output and a release-runtime valid-peer/blocked-peer trace; ordinary CI does not exercise the changed release selector or establish rejection before authenticated HTTP effects.
Agent review details

How this fits together

GitHub release automation selects Node, installs dependencies, validates the package, and publishes through npm trusted publishing; this change affects that tooling runtime.

flowchart TD
  A[Release trigger] --> B[Setup Node 24.22.1]
  B --> C[Install dependencies]
  C --> D[Validate package]
  D --> E[npm trusted publishing]
  E --> F[GitHub release]
Loading

Technical review

Best possible solution:

Keep the existing release workflow owner and exact patched-runtime selection, with inspectable evidence of valid-peer access and compromised-key rejection.

Do we have a high-confidence way to reproduce the issue?

Main uses the broad Node 24 selector, which can select an older cached patch, and upstream source establishes the missing compromised-key rejection; no supplied runtime trace establishes that release jobs actually selected 24.21.0.

Is this the best way to solve the issue?

An exact security-patch selection through the existing NODE_VERSION input is a small, appropriate fix that preserves package engines, router behavior, CI majors, and publication permissions.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 3db9a80449c1.

Merge-risk options

Maintainer options:

  1. Attach release-runtime TLS evidence (recommended)
    Provide redacted terminal output or a linked artifact showing Node 24.22.1 in the release-tooling environment, a successful valid-peer request, and rejection of a compromised-key chain before an authenticated HTTP request reaches the peer.

Provenance checked

  • Release Node version selection changes intended behavior with a stated reason (61fb4ea: The Node 24 selector was introduced as part of npm trusted-publishing automation.)

Testing

Proof path: unit tests only.

Security

Needs attention: The patch adopts upstream TLS key revocation without expanding workflow permissions, but its authority-boundary evidence remains incomplete.

Evidence

What I checked:

  • Verified introduced change: The pinned introduction diff changes only NODE_VERSION from 24 to 24.22.1 and adds the release-tooling changelog entry; main retains 24. (.github/workflows/release.yml:26, cd6172b21a13)
  • Release ownership and original intent: Release automation originally introduced the Node 24 selector with npm trusted publishing; subsequent work added non-publishing validation in ci: move macOS checks to Xcode 27 runners #63. (.github/workflows/release.yml:26, 61fb4ea24131)
  • Authoritative runtime security contract: The workflow executes Node, so its TLS contract directly governs release-tooling HTTPS requests. The upstream change blocks eight compromised eMudhra/emSign intermediate keys during certificate verification, reports CERT_REVOKED, and preserves rejectUnauthorized semantics. The Node 24.22.1 release notes include this commit. (src/crypto/crypto_context.cc, 01acf0ee7493)
  • Version-selection boundary: The pinned action accepts exact version specifications and defaults check-latest to false; its own action runtime remains separately selected by the provider. (action.yml, 949feb2413d6)
  • Discussion and supplied evidence: The full body claims independent incident-chain qualification but supplies no trace or linked artifact; both discussion comments contain review-status information and there are no submitted reviews. (cd6172b21a13)
  • Validation and policy: Root AGENTS.md was read in full; no nested policy, maintainer notes, or pr-simplify skill was present. YAML parsing and diff whitespace checks pass, and exact-head checks report twelve successes and two skipped dispatch jobs. No target code was executed and the checkout remains clean. (AGENTS.md, cd6172b21a13)

Likely related people:

  • Peter Steinberger: Raw commit 3b2fb01 adds .github/workflows/release.yml:30 relative to its recorded parents. This identifies author metadata, not feature responsibility or a PR merger. (role: source-line author; confidence: high; commits: 3b2fb01297a5; files: .github/workflows/release.yml)

Labels

Label changes:

  • add P2: This is a focused release-tooling security improvement; exposure in an actual release run has not been demonstrated.
  • add merge-risk: 🚨 security-boundary: The selected runtime changes TLS peer authorization for dependency and registry requests, and final-effect evidence is not attached.
  • add rating: 🧂 unranked krab: Overall readiness is 🧂 unranked krab; proof is 🧂 unranked krab and patch quality is 🦐 gold shrimp.
  • add status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask.

Label justifications:

  • P2: This is a focused release-tooling security improvement; exposure in an actual release run has not been demonstrated.
  • merge-risk: 🚨 security-boundary: The selected runtime changes TLS peer authorization for dependency and registry requests, and final-effect evidence is not attached.
  • rating: 🧂 unranked krab: Overall readiness is 🧂 unranked krab; proof is 🧂 unranked krab and patch quality is 🦐 gold shrimp.
  • status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask.

Rank-up moves

Optional improvements that raise the rating; they are not merge blockers.

  • Add final-effect proof that a valid TLS peer remains usable and a compromised-key peer receives no authenticated HTTP request under the selected release runtime.

Rating scale

6/6 🦀 challenger crab · 5/6 🦞 diamond lobster · 4/6 🐚 platinum hermit · 3/6 🦐 gold shrimp · 2/6 🦪 silver shellfish · 1/6 🧂 unranked krab. Overall follows the weaker of proof and patch quality; ✨ marks media proof (a screenshot, video, or linked artifact) that directly shows the changed behavior.

Workflow

ClawSweeper edits this one comment on every review. Comment @clawsweeper re-review for a fresh review only. Repair and merge execution lanes are retired.

Reviewed October 21, 2026, 6:22 PM ET / 22:22 UTC.

@joshavant
joshavant merged commit 33ea5f3 into main Oct 22, 2026
20 checks passed
@joshavant
joshavant deleted the security/emudhra-node-runtime branch October 22, 2026 17:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

merge-risk: 🚨 security-boundary 🚨 Merging this PR could weaken sandboxing, authorization, credentials, or sensitive data. P2 Normal priority bug or improvement with limited blast radius. rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant