The Agent Skills Directory

[PROMPT_INJECTION]: The skill authorizes the use of Read, Grep, and Glob tools to analyze local project files, creating a potential surface for indirect prompt injection. * Ingestion points: File content is ingested via the Read, Grep, and Glob tools. * Boundary markers: There are no instructions defining explicit delimiters or guidance to ignore commands embedded within the processed data. * Capability inventory: The skill has access to Bash, Write, and Edit tools, which could perform sensitive operations if an injection is successful. * Sanitization: The skill does not implement validation or filtering of the content read from external files before processing.

microservices-patterns