The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing subcommands of the ae-cli tool to perform data analysis and retrieval tasks. This is the primary intended functionality of the skill.
[PROMPT_INJECTION]: The skill processes untrusted, externally-sourced data from community platforms (such as posts and comments) which constitutes an indirect prompt injection surface. This is an inherent aspect of the skill's purpose as an analysis tool.
Ingestion points: Untrusted data is ingested through subcommands like +search_posts and +get_post_detail, specifically as documented in references/community-analyzing-theme-comment.md and references/community-character-analysis.md.
Boundary markers: The instructions do not define explicit boundary markers or delimiters for the ingested community text to isolate it from the agent's instructions.
Capability inventory: The skill possesses the capability to execute shell commands via the ae-cli binary.
Sanitization: While the skill provides instructions to filter data for quality (e.g., removing short comments or spam), there are no specific security-focused sanitization steps to prevent command or prompt injection from the community data itself.

ae-community