The Agent Skills Directory

[SAFE]: The skill is entirely instructional, providing metrics calculation logic and industry benchmarks. It does not request or execute any shell commands, scripts, or remote resources.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied post metrics and comments. While this presents an ingestion surface, the risk is negligible as the skill lacks access to any tools or capabilities that could be exploited (e.g., no network operations or file system modifications). Boundary markers and data sanitization are not explicitly defined, but are mitigated by the absence of execution capabilities. 1. Ingestion points: User-provided content for Post metrics in the Information Gathering section. 2. Boundary markers: Absent. 3. Capability inventory: No file system writes, network operations, or shell command executions are defined. 4. Sanitization: Absent.

engagement-rate-calculator-benchmarker