Agent Evolver

• Analyzed: May 15, 2026, 4:50 AM
• Findings: 1
• Version: 1.0.0
• Hash: fb2795eb441d9affc0f05309e1310fa4b26afc0895495e752ccfcbc961d52813
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill bundle provides an AI agent self-evolution engine, utilizing LLMs for error analysis and a local SQLite/ChromaDB for experience storage. The `deploy.sh` script performs standard installation tasks like creating local data directories (`~/.evolver`), installing Python dependencies from `requirements.txt`, and copying skill files to the OpenClaw workspace. Python scripts handle arguments via `argparse` and interact with OpenAI APIs using `requests` for LLM and embedding services, which is consistent with its stated purpose. While the LLM integration presents a potential prompt injection vulnerability against the LLM itself, there is no evidence of intentional malicious behavior such as data exfiltration to unauthorized endpoints, unauthorized remote code execution, persistence mechanisms outside of standard skill installation, or malicious prompt injection against the OpenClaw agent to perform harmful system actions. All file operations and external calls are aligned with the skill's described functionality.