The Agent Skills Directory

[SAFE]: The skill is a pure prompt-based interviewer. It does not use any tools (allowed-tools is empty) and does not perform any network operations or execute external code.
[SAFE]: It emphasizes user privacy by instructing the agent to explicitly ask for permission regarding naming clients or companies and to respect user limits ('I would rather not say').
[SAFE]: The skill includes a specific section for handling untrusted content, instructing the agent to treat pasted text or external data as information rather than instructions, which mitigates indirect prompt injection risks.
[SAFE]: The skill's primary function is to write a local file (story-bank.md) based on the conversation. It explicitly warns the user that this file should be gitignored to prevent accidental exposure of personal data in version control.

linkedin-interviewer