catalog

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). Several URLs point to direct installer scripts (.sh), downloadable archives, and a generic GitHub releases page from untrusted/example domains—common vectors for distributing malware if executed without verification.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The repository instructs agents to dynamically fetch and load external skill instructions and offers features (e.g., sendDataModel) that attach the entire client data model to outbound messages and npx-based installs, creating high-risk supply-chain and data-exfiltration vectors if abused or misconfigured.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). This skill explicitly instructs the agent at runtime to fetch and load external skill instructions from https://skills.danicat.dev///SKILL.md (and related index files like https://skills.danicat.dev/llms.txt), which are then injected into the active context and directly control agent behavior.