The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes a code review feature where users are encouraged to paste code snippets for the AI to analyze. This creates a potential surface for indirect prompt injection if the user-provided code contains instructions designed to subvert the agent's logic.
Ingestion points: Code snippets provided by users in response to the 'Code Review' instructions in SKILL.md and SKILL_v1.1.md.
Boundary markers: Absent; the templates do not include delimiters or instructions for the agent to disregard instructions embedded within the code being analyzed.
Capability inventory: The skill does not include standalone executable scripts or tools, but it leverages the agent's general reasoning and file analysis capabilities.
Sanitization: Absent; there are no instructions for the agent to validate or sanitize the incoming code for potential injection strings.

ai-learning-planner-cn