memory-lancedb-pro
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill implements an automated memory retrieval system that injects past conversation data into the agent's context, which presents a surface for indirect prompt injection. Ingestion points: Found in the autoRecall hook and memory_recall tool defined in SKILL.md. Boundary markers: Includes instructions for the agent's system prompt to treat memories as internal reference and avoid quoting them directly (SKILL.md). Capability inventory: The skill executes openclaw CLI commands and performs network operations to AI providers (SKILL.md). Sanitization: Employs LLM-powered extraction and noise filtering logic to manage the quality of stored and retrieved data (references/full-reference.md).
- [EXTERNAL_DOWNLOADS]: Downloads plugin components and configuration templates from the vendor's GitHub repository (github.com/CortexReach) and uses the openclaw plugin manager to install packages.
- [COMMAND_EXECUTION]: Utilizes local CLI commands like curl and openclaw to verify API credentials and manage the local configuration file (openclaw.json). These checks are directed at well-known service endpoints and the local environment.
Audit Metadata