Agent Skills Search

• Analyzed: May 14, 2026, 3:10 AM
• Findings: 1
• Version: 1.0.0
• Hash: 811f9748f00493d1ff166ca9ce10927bdda4d5ef24a25c3929a5707448f0cbe6
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill bundle is classified as benign. The `SKILL.md` instructs the agent to execute a local Python script (`search.py`) with a user-provided query. The `search.py` script safely handles the query as a string for filtering purposes, making an expected network request to `https://agnxi.com/sitemap.xml` to fetch the sitemap. There is no evidence of intentional malicious behavior such as data exfiltration, malicious execution within the script, persistence mechanisms, or prompt injection attempts against the agent to deviate from its stated purpose. The shell execution instruction in `SKILL.md` is the intended mechanism for the skill to function, and the script itself processes input securely.