The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and structured templates. It does not include any scripts, binary files, or configuration that enables command execution or network access.- [PROMPT_INJECTION]: The skill uses slash commands like '/positioning-audit' as references to related skills within the agent environment. These are used for workflow navigation rather than bypassing safety protocols. No adversarial injection patterns were found.- [DATA_EXFILTRATION]: No hardcoded credentials, sensitive file paths, or network exfiltration patterns were detected in the instructions or metadata.- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided messaging documents for auditing. While this represents a data ingestion surface, the skill lacks high-privilege capabilities such as file writing or network requests, preventing potential data-driven instructions from causing system-level harm.

messaging-hierarchy