The Agent Skills Directory

[SAFE]: The skill's primary function is to facilitate a discovery process through structured natural language interaction. All instructions are focused on logical consistency, ambiguity reduction, and specification architecture.
[COMMAND_EXECUTION]: The skill utilizes shell commands (tldr structure, tldr arch, tldr tree, tldr calls) to inspect the local filesystem. This is a legitimate functional requirement to support 'brownfield' project analysis, allowing the agent to understand existing technology stacks and patterns before starting the interview.
[DATA_EXPOSURE]: The skill accesses project directory structures and source code files to inform its reasoning. This data is used locally to refine the interview process and is not transmitted to external domains or third-party services.

deep-interview