Security audit
Auto Session Labels
Security checks for vulnerabilities and agentic risk
Overview
This plugin does what it says: it automatically uses an LLM to title recent unlabeled conversations, with scoped controls and no evidence of hidden or destructive behavior.
Install only if you are comfortable with recent conversation snippets being sent to the agent's default or configured model provider to generate labels. Administrators should review the model setting, consider dryRun for testing, and update development dependencies before local development or CI use.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
No suspicious patterns detected.