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.
Warn5 providerslatest audit Sep 15, 2026
Independent checks from skills.sh's audit partners.
This skill allows AI agents to log errors, user corrections, and insights into project memory files to improve future performance. It includes scripts for automatic error detection and skill scaffolding. The primary security consideration is indirect prompt injection, where untrusted data from tool outputs or user feedback is stored in instructions that the agent treats as authoritative for future sessions.
Detected behaviors
No alerts
1 issue
4/11 files flagged
Score: 93/100 · 2 sections analyzed