obsidian-brain
Fail
Audited by Snyk on Aug 3, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E005: Suspicious download URL detected in skill instructions.
- Suspicious download URL detected (high risk: 0.70). The GitHub URL points to a third‑party personal repo that the skill instructs users to clone/install (allowing arbitrary unvetted code and scripts to run), which is potentially risky, while obsidian.md is the official vendor site and not suspicious.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The required workflow reads free text from outsider-authored sources only when the user imports them into the local vault (e.g., emails, WhatsApp chats, and transcripts are ingested into
sources/and then routed for grounding via the read-first protocol), meaning the agent can consume attacker-controlled text written into those user-provided inputs.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill's install instructions tell users to fetch a remote git repo that provides the agent skills (which directly control agent instructions) via "git clone --depth 1 https://github.com/Thibaultbm/claude-seo-geo.git", so runtime installation pulls external code that determines agent behavior.
Issues (3)
E005
CRITICALSuspicious download URL detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata