The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructs the agent to update environment variables in .env files and configuration secrets. However, it explicitly mitigates risk by instructing the agent to 'flag secret rotation as the user's responsibility' and 'don't invent new key values,' which is a security best practice for migration tasks.
[INDIRECT_PROMPT_INJECTION]: The skill operates by reading and analyzing existing user source code (via Read, Grep, and Glob tools). While this represents a data ingestion surface where malicious instructions could theoretically be embedded in the code being analyzed, the skill's logic is focused on static mapping and transformation, minimizing the risk of accidental execution of embedded instructions.
[COMMAND_EXECUTION]: The skill uses Bash(pip *) and Bash(npm *) to update project dependencies. This is a necessary and expected capability for a migration tool to ensure the Tavily SDK is correctly installed in the user's environment.
[METADATA_POISONING]: The skill's metadata (name, description, and allowed tools) is consistent with its stated purpose of providing a migration path for developers switching to the Tavily platform.

migrate-to-tavily