Security audit
Human Style Writing
Security checks for vulnerabilities and agentic risk
Overview
This is a Markdown-only writing helper for making chat and social posts sound more natural, with no code execution, credentials, or automatic account actions.
Use this as a drafting and tone aid, not to impersonate someone, fabricate personal experiences, or hide AI involvement where disclosure is expected. If you reuse chat or writing samples, remove names, private details, and sensitive context first.
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.