turso
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches official installation scripts and CLI binaries from Turso's official domain (get.tur.so) and its official GitHub repository.
- [REMOTE_CODE_EXECUTION]: The documentation provides commands to download and immediately execute installation scripts using shell pipes (bash/sh) or PowerShell's Invoke-Expression (iex).
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill is designed to manage agent state, memory, and history in libSQL databases, as outlined in
references/agents.md. - Boundary markers: The documentation does not specify explicit boundary markers or instructions to delimit untrusted data from system instructions within the database schema.
- Capability inventory: The skill uses libSQL drivers which perform local file system operations (SQLite file creation/updates) and network synchronization with the Turso Cloud platform.
- Sanitization: Code examples demonstrate the use of prepared statements with positional parameters (
?) to prevent SQL injection, though this does not mitigate higher-level prompt injection from retrieved data.
Audit Metadata