Project Risk Score Project Risk Score: weighted blend of vulnerability (40%), trust (30%), and pipeline (30%) subscores. Useful for at-a-glance triage. Click for more →
61 /100
Review required

4 findings across 1 packages · 1412 packages clean

Candidates: 2 auto-merge · 0 review · 0 decline · 0 no fix

Scanned webpack/webpack @ main Scan · Completed 215 hrs ago Download SBOM
Total Findings
4
All detected issues
KEV Findings
0
Known exploited
High EPSS
0
Likely to be exploited
Auto-merge ready
2
Remediation candidates
Affected pkgs
1
with remediation paths
KEV catalog CISA's Known Exploited Vulnerabilities catalog. Documented active exploitation in the wild; the strongest 'this is happening now' signal. Click for more →
0
actively exploited CVEs
Highest EPSS Exploit Prediction Scoring System. Daily-updated probability that a CVE will be exploited in the next 30 days. Probability, not severity. Click for more →
0.4%
CVE-2026-13149
Active vetos
0
lens blocks on candidates

Findings

brace-expansion @ 1.1.15
root → eslint → minimatch → brace-expansion
4 findings medium
Trust 0/100 EPSS 0.4%
AUTO-MERGE Verdict tier: the fix passes all three lenses cleanly. In Mode C, Arguss opens a pull request for the upgrade; it does not merge PRs on GitHub. Click for more →
GHSA-3jxr-9vmj-r5cp medium CVSS 5.3 EPSS 0.4%
brace-expansion: DoS via exponential-time expansion of consecutive non-expanding {} groups

Affects 4 install paths

1.1.15 → 1.1.16 patch
auto-merge Verdict tier: the fix passes all three lenses cleanly. In Mode C, Arguss opens a pull request for the upgrade; it does not merge PRs on GitHub. Click for more → Score 100
Max CVSS 5.3 · 1 finding Common Vulnerability Scoring System. A 0–10 score for how damaging exploitation could be. Severity, not urgency. Click for more → Max EPSS 0.4% · 28th percentile Exploit Prediction Scoring System. Daily-updated probability that a CVE will be exploited in the next 30 days. Probability, not severity. Click for more →
  • patch-level upgrade; trust signals unchanged; CI verifies tests
GHSA-3jxr-9vmj-r5cp: brace-expansion: DoS via exponential-time expansion of consecutive non-expanding {} groups

Upgrade brace-expansion from 1.1.15 to 1.1.16 or later

View advisory
root → eslint → minimatch → brace-expansion
2.1.1 → 2.1.2 patch
auto-merge Verdict tier: the fix passes all three lenses cleanly. In Mode C, Arguss opens a pull request for the upgrade; it does not merge PRs on GitHub. Click for more → Score 100
Max CVSS 5.3 · 3 findings Common Vulnerability Scoring System. A 0–10 score for how damaging exploitation could be. Severity, not urgency. Click for more → Max EPSS 0.4% · 28th percentile Exploit Prediction Scoring System. Daily-updated probability that a CVE will be exploited in the next 30 days. Probability, not severity. Click for more →
  • patch-level upgrade; trust signals unchanged; CI verifies tests
GHSA-3jxr-9vmj-r5cp: brace-expansion: DoS via exponential-time expansion of consecutive non-expanding {} groups

Upgrade brace-expansion from 2.1.1 to 2.1.2 or later

View advisory
root → eslint → minimatch → brace-expansion

Blast radius

Paths from project root to brace-expansion - which dependencies pulled this package in?

Package status

1414 packages scanned.

  • 2 Auto-merge candidates

Review auto-merge candidates and open PRs in a guided flow.

Glossary

What the labels and signals mean.
Trust Save
A package upgrade Arguss would have blocked despite the new version being available, because trust signals, like ownership transfer or a new maintainer, fired during the upgrade window. The name reflects what the agent did for the user: saved them from a potentially malicious update that a version-only auto-PR tool would have merged.
AUTO-MERGE
Verdict tier indicating the fix passes all three lenses cleanly. After you confirm action from a Scan assessment, Arguss opens a pull request for the upgrade; it does not merge PRs on GitHub. The envelope is conservative on purpose: patch or minor version bump, trust signals unchanged, blast radius bounded, real tests pass.
REVIEW
Verdict tier requiring a human decision. At least one veto fired during fix-confidence evaluation, trust signals shifted, the pipeline can't verify post-upgrade behavior, or the upgrade is a major version bump. The agent surfaces the reasons; the developer decides.
DECLINE
Verdict tier indicating no remediation is recommended. Typically applies when no fix version exists for the finding, or when multiple critical vetoes make even human review unproductive.
fix_kind.major
Veto signal that fires when the available fix requires a major version bump (1.x → 2.x). Major bumps imply potential breaking changes and fall outside the auto-merge envelope by default, even when the upgrade is the only available fix.
trust.new_maintainer
Veto signal that fires when a package added a new maintainer during the upgrade window, meaning between the user's current version and the proposed upgrade. New publishing identities are a well-documented attack vector for typosquats and supply chain takeovers.
trust.ownership_transferred
Veto signal that fires when a package's primary maintainer changed during the upgrade window. Combined with trust.new_maintainer, this is the highest-risk trust combination, typical of the xz-utils style attacks and historical npm credential theft incidents.
pipeline.test_reality
Veto signal that fires when Arguss can't verify tests will run on the upgraded code. Four conditions must hold: a test script exists in package.json, it isn't a no-op, real test files exist, and a workflow actually invokes them. If any fail, the fix cannot qualify for AUTO_MERGE because there's no way to verify the upgrade didn't break the user's project.
CVSS
Common Vulnerability Scoring System. A numeric score (0.0–10.0) representing how damaging a vulnerability could be if exploited. Sourced from NIST's National Vulnerability Database via OSV.dev. Severity, not urgency.
EPSS
Exploit Prediction Scoring System. A daily-updated probability (0.0–1.0, displayed as percent) that a CVE will be exploited in the next 30 days. Sourced from FIRST.org. Probability, not severity.
KEV
CISA's Known Exploited Vulnerabilities catalog. A federal list of CVEs with documented active exploitation in the wild. Federal agencies have a binding patching deadline; for everyone else, presence on KEV is the strongest "this is being used right now" signal available. Sourced directly from CISA.
Project Risk Score (PRS)
A weighted blend of the three lens subscores (40% vulnerability, 30% trust, 30% pipeline) producing an overall 0–100 indicator of the project's dependency health. Useful for at-a-glance triage; the per-finding fix-confidence verdicts are what drive automated decisions.

Dependency graph

Full-project map of transitive dependencies. Severity colors reflect vulnerabilities; trust rings apply only to direct dependencies analyzed by OpenSSF Scorecard (higher = riskier).