self-eval
Automatically evaluates session outcomes by checking for errors and logs reflections to LanceDB, enhancing AI learning.
Install this skill
or
95/100
Security score
The self-eval skill was audited on May 29, 2026 and we found 1 security issue across 1 threat category. Review the findings below before installing.
Categories Tested
Security Issues
medium line 40
Access to hidden dotfiles in home directory
SourceSKILL.md
| 40 | - 读取 `~/.openclaw/agents/main/sessions/*.jsonl` |
Scanned on May 29, 2026
View Security Dashboard