The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides templates and patterns for the agent to analyze and rewrite user-supplied code, database queries, and API response structures. This processing of potentially untrusted data constitutes a surface for indirect prompt injection.
Ingestion points: The agent ingests external code and data for optimization through templates like templates/performance-test.js and examples in examples/.
Boundary markers: The provided templates and instructions do not include explicit boundary markers or instructions to the agent to disregard potential hidden instructions within the data being analyzed.
Capability inventory: According to the allowed-tools in SKILL.md, the agent has access to Write, Bash, and MultiEdit tools, which increases the potential impact of a successful injection.
Sanitization: There is no specific evidence of input sanitization or validation logic within the skill's instructions for the data it processes.

performance-optimization