The Agent Skills Directory

[SAFE]: The skill consists of natural language instructions for an AI agent's internal workflow. It does not include any executable code, shell commands, or network operations.
[DATA_EXFILTRATION]: No network access or data transmission patterns were identified. The skill operates entirely within the agent's context.
[REMOTE_CODE_EXECUTION]: The skill does not download external scripts or packages. It relies on previously defined internal skills like 'app-knowledge' and 'resolve-ambiguity'.
[PROMPT_INJECTION]: The instructions do not contain any patterns designed to bypass safety filters or override system constraints. The workflow focus is on task clarification and planning.

coding-init