status
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill accesses the sensitive file path
~/.pi/agent/mem0-config.jsonand theMEM0_API_KEYenvironment variable to verify configuration. It also instructs the agent to display the first six characters of the API key in the status output, which constitutes a partial credential exposure. - [INDIRECT_PROMPT_INJECTION]: The skill processes data retrieved from the
mem0_memorytool during connectivity and quality checks, creating a surface for indirect prompt injection if the stored memories contain malicious instructions. - Ingestion points: Data is ingested from the
mem0_memorytool viasearchandget_allactions as described inSKILL.md. - Boundary markers: No delimiters or instructions to ignore embedded content are specified for the memory data processing steps.
- Capability inventory: The skill utilizes the
mem0_memorytool for adding, searching, and retrieving memories. - Sanitization: There is no mention of sanitization, validation, or filtering of the retrieved memory content before it is processed by the agent for quality analysis.
Audit Metadata