The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface where user-supplied inputs are incorporated into the agent's core instructions.
Ingestion points: The topic_or_problem, context, prior_sessions, and developmental_band fields defined in the input_schema of SKILL.md are interpolated directly into the system prompt.
Boundary markers: The prompt uses plain-text labels and horizontal rules to separate user data from instructions, but lacks strong delimiters (like XML tags or triple quotes) or specific guardrail language to prevent the agent from following instructions embedded within the user data.
Capability inventory: There are no dangerous capabilities present; the skill has no access to the shell, file system, network, or external tools.
Sanitization: No input validation or sanitization is performed on the user-provided strings.
[NO_CODE]: This skill contains no executable scripts, binaries, or logic beyond the system prompt and markdown instructions.

explain-first-interrogator