Dynamic code execution
Critical
- Finding
- Dynamic code execution detected.
Agent Architecture Patterns
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill is mostly an educational agent-pattern library, but a runnable ReAct example/test uses JavaScript eval for calculator input, which can execute arbitrary local code if reused with untrusted expressions.
This appears to be an educational pattern library rather than a credentialed or persistent integration. Before installing or running examples, be aware that the ReAct calculator example uses JavaScript eval; do not run or reuse it with untrusted input unless you replace it with a safe math parser. Also consider the lack of a verified source/homepage when deciding whether to trust updates.
Static analysis
Dynamic code execution
Critical
- Finding
- Dynamic code execution detected.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
What this means
If a user runs or reuses this example with untrusted input, a crafted expression could access local files or run code with the user's local Node.js privileges.
Why it was flagged
The ReAct calculator tool runs a string supplied as a tool argument. In agentic use, that argument may be generated from user/model input, so it can execute arbitrary JavaScript in the Node.js process rather than only math.
Skill content
execute: async ({ expression }) => { ... const result = eval(expression);Recommendation
Replace eval with a strict arithmetic expression parser or an allowlist of permitted operators, and do not expose this tool to untrusted or model-generated expressions.
What this means
It is harder for a user to verify the package origin, review upstream changes, or confirm maintainership.
Why it was flagged
The skill includes runnable code examples, but the registry metadata does not provide an authoritative source or homepage for independent verification.
Skill content
Source: unknown; Homepage: none
Recommendation
Review the included files before use and prefer installing skills that provide a verifiable source repository or homepage.