The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and templates for content generation. It does not request network access, file system permissions, or the execution of external scripts.
[INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface as it ingests user-provided topics, context, and external audit data to generate posts. However, it lacks exploitable capabilities (such as command execution or network exfiltration), meaning any injected instructions would only affect the text output provided to the user.
Ingestion points: Topic, Pillar, and Context inputs in SKILL.md; Audit data in the GEOTOOL integration section.
Boundary markers: Absent.
Capability inventory: None. The skill only performs text generation within the LLM context.
Sanitization: Absent.

linkedin-post