The Agent Skills Directory

[SAFE]: The skill facilitates a structured interview to generate a local configuration file (~/.claude/company-context.md). This behavior is consistent with its stated purpose of providing context to an AI agent acting as a business advisor.
[DATA_EXPOSURE]: The skill is designed to collect and store highly sensitive business intelligence, including revenue figures, growth rates, monthly burn, and strategic risks, in a plain-text markdown file. While this is the primary function of the skill, users should be aware that this sensitive information is consolidated in a predictable local file path.
[INDIRECT_PROMPT_INJECTION]: The skill represents an ingestion surface as it captures user input and writes it to a file that other skills (like context-engine or cs-chief-of-staff) are intended to read.
Ingestion points: User responses to the 12-question interview (SKILL.md).
Boundary markers: The output uses structured markdown headers and bullet points, but there are no specific instructions to ignore embedded commands in the user's input.
Capability inventory: The skill performs file system writes to ~/.claude/company-context.md and suggests the use of symlinks (ln -sf).
Sanitization: No explicit sanitization or validation of the user's input is performed before writing it to the durable context file.

onboard