Memberstack CLI

• Verdict: Pass
• Analyzed: May 12, 2026, 4:16 AM
• Findings: 0
• Version: 1.1.0
• Hash: 91e793ab2a0f0dc132434d8c6d063b09c599179c715c989a98cc46ae4db7fa7e
• Source: palm
• External report: View on VirusTotal

Overview

Type: OpenClaw Skill Name: memberstack-cli Version: 1.1.0 The skill bundle demonstrates a strong commitment to security, particularly against prompt injection and sensitive data exposure in an AI agent context. The `SKILL.md` provides explicit instructions to the agent to sanitize output, use boundary markers, confirm destructive actions, and never access or log credentials. The `scripts/run_memberstack.py` wrapper enforces these rules by implementing output sanitization, preventing direct access to sensitive credential paths, and requiring confirmation for destructive commands. The `scripts/search_references.py` also sanitizes documentation content. There is no evidence of intentional harmful behavior, data exfiltration, or backdoors; instead, the code actively mitigates common security risks.