The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to install the Turso CLI by piping a remote shell script directly into the shell (curl | sh). This script is sourced from the official repository of the database provider.
[EXTERNAL_DOWNLOADS]: Fetches installation assets and configuration scripts from the official Turso Database GitHub releases and installs Node.js dependencies from the NPM registry.
[COMMAND_EXECUTION]: The skill requires executing various shell commands to function, including curl for installation, brew for package management, and the tursodb CLI for database interactions.
[INDIRECT_PROMPT_INJECTION]: The skill operates on database content (such as through full-text search and vector similarity queries) which can originate from untrusted external sources. This creates a surface where embedded instructions in the data could potentially influence the agent's logic during retrieval operations.
Ingestion points: Database tables, query results from the tursodb CLI, and FTS/Vector search outputs in SKILL.md and references/full-text-search.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided in the skill files.
Capability inventory: The skill can execute shell commands (tursodb), write to the local file system (database files), and perform network operations (installation).
Sanitization: There are no documented mechanisms for sanitizing or validating database content before it is processed by the agent.

turso-db