The Agent Skills Directory

[SAFE]: The skill provides a structured template and workflow for documentation synthesis. It does not exhibit any malicious patterns, obfuscation, or unauthorized access to sensitive system resources.
[DATA_EXPOSURE]: The skill manages local documentation files in paths like work/ and docs/agents/issue-tracker.md. This behavior is consistent with its stated purpose of specification publication and project documentation management.
[COMMAND_EXECUTION]: The skill references internal agent commands (e.g., /council, /best-practices-research, /setup-work, /to-tickets) to sequence validation and setup tasks. These are platform-specific instructions for the agent's internal workflow and do not involve arbitrary shell execution or remote code execution.
[PROMPT_INJECTION]: The instructions do not contain any patterns designed to override safety filters or bypass agent constraints. The inclusion of disable-model-invocation: true in the frontmatter serves as an additional safety restriction on the skill's capabilities.

to-spec