The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions create a surface for indirect prompt injection by directing the agent to ingest and follow instructions found within documentation files stored in the .claude/solutions/ directory. * Ingestion points: Solution files located within the .claude/solutions/ directory and its subdirectories. * Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when the agent processes the content of these documentation files. * Capability inventory: The skill utilizes standard tools like Grep and Glob for searching and integrates with core agent workflows such as planning and execution (/rb:plan and /rb:work). * Sanitization: There are no requirements or methods specified for sanitizing or validating the content of the solution documents before they are analyzed by the agent.

compound-docs