The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted data from external sources.
Ingestion points: Data enters the agent's context through the memory_search tool and user responses collected via ask_user_question as described in §0 and §7.2.
Boundary markers: The skill lacks explicit syntax delimiters (e.g., XML tags or unique markers) for the injected data, though §17 provides strong instructional constraints ('Hard Floors') to prevent the agent from obeying fabricated instructions or facts.
Capability inventory: The skill possesses the capability to modify the agent's long-term state via memory_save and interact with the user via ask_user_question based on processed data.
Sanitization: There is no evidence of sanitization, validation, or filtering of the content retrieved from memory before it is interpolated into the drafting process.

writing-style