The Agent Skills Directory

[SAFE]: The skill instructions focus on qualitative research synthesis and business decision-making. No prompt injection, obfuscation, or persistence mechanisms were detected.
[NO_CODE]: The skill is composed entirely of markdown instructions and YAML configuration metadata. It does not include or execute any scripts, binaries, or external code.
[DATA_EXFILTRATION]: The skill manages local business context and evidence files within the .business/ directory. It includes explicit privacy safeguards, instructing the agent to obtain consent before storing personal information and to avoid importing private records without confirmation. No network capabilities or exfiltration patterns are present.
[PROMPT_INJECTION]: The instructions do not contain attempts to bypass safety filters, override system behavior, or extract sensitive instructions. It follows standard agent task modeling.

understand-customers