Dialogflow Cx Agents

• Analyzed: May 11, 2026, 4:57 AM
• Findings: 1
• Version: 1.0.0
• Hash: 91412b0ebba43841796aaa33037825811d098d765ccc8def51467d1e98223093
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill bundle is classified as suspicious due to potential vulnerabilities rather than explicit malicious intent. The `scripts/agents.py` file includes an `export_agent` function that writes to a user-specified `output_file`. If the OpenClaw agent or user input does not properly sanitize this argument, it could lead to arbitrary file write (e.g., path traversal), allowing an attacker to overwrite or create files in arbitrary locations. Additionally, the `SKILL.md` provides `curl` examples that, if executed directly by an agent with unsanitized user input, could be vulnerable to shell injection. While these are significant risks, they represent vulnerabilities that *allow* attacks rather than code *designed* for malicious actions like data exfiltration to unauthorized endpoints or backdoor installation.