The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses clear persona-defining instructions that align with standard AI assistant usage. There are no attempts to bypass safety filters or override system-level instructions.
[DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or network exfiltration patterns were identified.
[REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads, package installations, or remote script execution.
[OBFUSCATION]: The content is written in plain text with no evidence of encoded strings, hidden characters, or homoglyph-based attacks.
[INDIRECT_PROMPT_INJECTION]: The skill requests external data ingestion (e.g., existing content for analysis and repurposing). While this is an injection surface, it is standard for content analysis tasks and the skill does not possess dangerous tools (such as file system writes or network operations) that could be exploited by such data.
Ingestion points: Discovery Question 7 and Example 3 request users to provide existing content or analytics data.
Boundary markers: None explicitly defined in the instructions.
Capability inventory: No tools or shell commands are utilized in this skill.
Sanitization: None present, as the analysis is purely conversational.

content-platforms