The Agent Skills Directory

[SAFE]: The skill provides comprehensive instructional content and reference documentation for brand voice management. It defines a formal contract for project documentation, covering terminology, facts, and surface-specific registers.
[COMMAND_EXECUTION]: The skill instructs the agent to run a project-local linter script (python3 docs/brand/lint.py). This is a standard development workflow practice for verifying documentation consistency and does not involve downloading or executing code from external, untrusted sources.
[INDIRECT_PROMPT_INJECTION]: The skill processes project-level documentation (such as foundation.md and strings.md) to audit brand voice. While this constitutes an ingestion surface for untrusted data, the skill's capabilities are restricted to documentation management and local script execution, with no associated network exfiltration or dangerous operations that would make this surface exploitable.

brand-voice