The Agent Skills Directory

[SAFE]: The skill consists entirely of process-oriented markdown instructions without any executable code, scripts, or network operations. The disable-model-invocation setting further restricts active execution during skill loading.
[INDIRECT_PROMPT_INJECTION]: The skill provides templates for processing external task tickets into markdown briefs. This is a standard ingestion surface where untrusted data could theoretically contain malicious instructions; however, the skill is purely instructional, provides structural boundaries (markdown templates), and does not automate any high-risk capabilities.

parallel-work