trixel-hermes

Warn

Audited by Socket on May 20, 2026

2 alerts found:

Securityx2
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill’s footprint is too broad for its stated helper role, and it encourages autonomous execution of local scripts, repo sync, and SMS sending without explicit approval boundaries. No confirmed malware or hostile exfiltration endpoint is shown, but opaque local scripts, global Git actions, and real-world messaging create medium-to-high security risk.

Confidence: 83%Severity: 72%
SecurityMEDIUM
scripts/git_sync.sh

No explicit embedded malware indicators are present in the script itself (it is a straightforward shell automation around Git). The primary risk is operational and supply-chain adjacent: it automatically stages (`git add .`), auto-commits, and pushes (`git push origin main`) all local repository content from hardcoded personal storage paths to the configured `origin/main` remotes, without validating remote trust, filtering staged files, or requiring user confirmation—creating a credible pathway for accidental or malicious data disclosure if sensitive files exist in those directories or if `origin` is misdirected.

Confidence: 62%Severity: 70%
Audit Metadata
Analyzed At
May 20, 2026, 12:22 PM
Package URL
pkg:socket/skills-sh/Sebfranklin%2Ffranklin-skills%2Ftrixel-hermes%2F@82714bbdcdd69aa065b65d29308cb3d32333d850
Security Audit — socket — trixel-hermes