The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external web sources which could potentially contain malicious instructions.
Ingestion points: Search results, content snippets, and raw page content retrieved via the tvly search command as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or warnings for the agent to ignore instructions embedded in search results.
Capability inventory: The skill's capabilities are restricted to the tvly command line tool within the Bash environment.
Sanitization: No specific content sanitization or validation steps are defined for the retrieved data before it enters the agent context.
[COMMAND_EXECUTION]: The skill uses the tvly CLI to perform its primary function.
The execution is scope-limited via the allowed-tools metadata, which restricts the Bash tool to only tvly commands, adhering to the principle of least privilege.
The skill includes clear instructions for handling authentication (tvly login) and environment-specific behaviors, such as avoiding interactive flows in unattended environments.

tavily-search