Back to skill
Skillv0.3.1
VirusTotal security
MindClaw · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 30, 2026, 4:59 AM
- Hash
- 436e60f4136d1ca59dfa20c257d6ddcee3445775544bfdb4c4f88067cf78cf11
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: mindclaw Version: 0.3.1 The MindClaw skill bundle appears benign. It provides persistent memory and knowledge graph capabilities for AI agents, integrating with OpenClaw and Claude Desktop via the Model Context Protocol (MCP). The code demonstrates security awareness, notably by redacting potential credentials during auto-capture (`src/mindclaw/capture.py`). File system modifications (e.g., writing to `~/.mindclaw/config.json`, `~/.openclaw/tools.json`, `MEMORY.md`) are explicitly for tool configuration and data synchronization, which are core, stated functions of an agent memory skill. Network activity is limited to local Ollama instances for semantic embeddings (`http://localhost:11434`). There is no evidence of data exfiltration, arbitrary code execution, or malicious prompt injection attempts in the documentation or code.
- External report
- View on VirusTotal