The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes and displays content from a local database (elements.db), specifically fields like user_intent and prompt_text. These fields contain data that may have been generated by or for an LLM, representing a potential indirect prompt injection surface.
Ingestion points: elements.db (tables generated_prompts, elements) accessed via the prompt_analyzer module.
Boundary markers: The skill uses markdown formatting and horizontal rules to separate data, which provides basic structural delimitation but no explicit safety instructions to the agent to ignore instructions within the data.
Capability inventory: The skill's capabilities are focused on data retrieval, analysis, and formatting. No dangerous subprocess calls, file writing, or network operations are present in the skill definition.
Sanitization: No explicit sanitization or validation of the ingested prompt text is mentioned in the logic.

prompt-analyzer