OverviewHistoryStatsSecurity
Security
Independent checks from skills.sh's audit partners.
Independent checks from skills.sh's audit partners.
Independent checks from skills.sh's audit partners.
Pass3 providerslatest audit Aug 15, 2026
Independent checks from skills.sh's audit partners.
The skill describes a workflow for managing an AI agent's memory, including searching for context at the start of a task and saving decisions during the process. It includes explicit advice to avoid saving secrets. A low-severity risk is identified regarding the attack surface for indirect prompt injection, as the skill directs the agent to ingest and act upon data retrieved from a persistent memory store.
Detected behaviors
No alerts
No issues