notebooklm-research

Warn

Audited by Socket on Jul 20, 2026

2 alerts found:

SecurityAnomaly
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill is broadly aligned with research automation, but it relies on an unofficial third-party client that receives a persistent Google session cookie, processes large amounts of untrusted external content, and supports autonomous sharing/posting. The main concern is trust and credential forwarding to non-Google tooling, not confirmed malware.

Confidence: 88%Severity: 79%
AnomalyLOW
scripts/pipeline.py

No direct evidence of classic malware (backdoor, credential theft, obfuscated payloads, eval/exec, or exfiltration) was found in this fragment. However, there is a meaningful supply-chain risk: it can execute an external CLI and notably runs a fallback via 'npx trend-pulse ...', which may download and execute npm packages at runtime. Additionally, user-supplied URLs are passed to NotebookLM for source ingestion without visible validation. Recommend removing or pinning the npx fallback, enforcing URL allowlists/scheme checks, and reviewing NotebookLM artifact download/write behavior.

Confidence: 72%Severity: 55%
Audit Metadata
Analyzed At
Jul 20, 2026, 10:21 PM
Package URL
pkg:socket/skills-sh/claude-world%2Fnotebooklm-skill%2Fnotebooklm-research%2F@dc38bd8106a2c5f23557fc03bf895c53fb19f67e
Security Audit — socket — notebooklm-research