The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown and reference samples for persona emulation. No malicious patterns, obfuscation, or persistence mechanisms were detected.
[NO_CODE]: No executable scripts (Python, Node.js, Shell) or binary files are included in the skill package.
[DATA_EXFILTRATION]: No network tools, hardcoded credentials, or sensitive file path access patterns were identified. The disable-model-invocation: true setting further restricts the execution environment.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: User-provided text passed through the argument-hint for drafting.
Boundary markers: Not explicitly defined in the prompt interpolation.
Capability inventory: None. The skill has no file-write, network, or subprocess capabilities.
Sanitization: None identified.
Note: While the skill processes untrusted input, the total absence of tool-use capabilities renders the injection surface non-exploitable.

ghost-writing