Novyx Memory

Enterprise-grade persistent memory for AI agents — rollback, audit trails, knowledge graph, governed actions, time-travel debugging, and 60+ commands coverin...

This skill will automatically send user messages and agent responses to an external Novyx API and persist them (auto-save/auto-recall). That is expected for a memory skill but can leak secrets or PII. Before installing: 1) Confirm you trust https://novyxlabs.com and the default API host (index.js defaults to https://novyx-ram-api.fly.dev); 2) Use a dedicated/test API key (not a high-privilege account) if you want to try it; 3) Consider disabling autoSave/autoRecall in configuration if you need manual control; 4) Note a minor doc mismatch: NOVYX_API_URL is supported by the code but not listed in SKILL.md metadata — if you prefer self-hosting, set NOVYX_API_URL to your endpoint; 5) Running the provided test (npm test) will create/delete memories on the configured account, so run tests with a disposable key. If you need a deeper assessment, provide network logs or confirm the official API host and privacy/security docs for Novyx.

Type: OpenClaw Skill Name: novyx-memory Version: 3.0.0 The novyx-memory skill bundle provides a comprehensive interface for a third-party persistent memory service. It implements over 60 commands for memory management, including semantic search, cryptographic audit trails, and 'time-travel' rollback features. The code (index.js) is a standard API client using axios to communicate with a legitimate hosting platform (fly.dev). While it automatically captures conversation context, this behavior is clearly documented in SKILL.md and README.md as the core functionality. No evidence of malicious intent, obfuscation, or unauthorized data exfiltration was found.