The Agent Skills Directory

[SAFE]: The skill's instructions and evaluation files were analyzed and no security threats were identified.
[DATA_EXFILTRATION]: The skill instructs the agent to read local context files such as ".agents/product-marketing.md" or "product-marketing-context.md". These are standard project-level files for marketing context and do not contain sensitive system credentials or private keys. No network operations capable of exfiltrating data are present.
[REMOTE_CODE_EXECUTION]: There is no evidence of remote script fetching, package installations, or dynamic code execution commands within the skill.
[PROMPT_INJECTION]: The skill defines an expert persona and provides community management playbooks without any attempts to bypass safety filters, override system instructions, or extract sensitive prompt data.
[COMMAND_EXECUTION]: No shell commands, privileged operations, or persistence mechanisms were detected in the instructions or metadata.

community-marketing