secondme-openclaw-key-memory

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [DATA_EXFILTRATION]: Accesses the local file {baseDir}/.credentials to retrieve an authentication token used for API requests to the vendor domain app.mindos.com. This is a standard authentication mechanism for the author's own infrastructure.
  • [PROMPT_INJECTION]: The skill ingests user-provided content for long-term storage, creating a surface for indirect prompt injection. • Ingestion points: Data enters the system via the content field in the memory insertion request. • Boundary markers: There are no explicit markers or instructions to isolate user content from system prompts. • Capability inventory: The skill is capable of external network communication for storage and retrieval operations. • Sanitization: No sanitization, validation, or filtering of the ingested content is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 03:42 AM