The Agent Skills Directory

[SAFE]: The skill operates as a high-level orchestration bundle for compliance-related tasks. It lacks executable code, scripts, or runtime commands. Analysis of the markdown instructions shows a strong emphasis on safety and error-handling, particularly in preventing the AI from hallucinating legal facts or overstepping its role. It explicitly mandates that the agent must cite official sources (verified during runtime) and mark all legal applicability questions as 'REVIEW REQUIRED'. No prompt injections, data exfiltration patterns, or persistence mechanisms were found. The 'companions' listed are internal skill references for capability scoping and do not represent external, unverifiable dependencies.

compliance-all