The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data to generate a 'Current reality' summary, creating a minor surface for indirect injection from the codebase content.
Ingestion points: The instructions direct the agent to read the project structure, README, recent commits, and existing configuration files (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands during the interpolation of project data into the final profile.
Capability inventory: The skill utilizes Read, Write, Glob, and Bash tools to perform project inspection and store configuration.
Sanitization: No specific filtering or validation of the ingested codebase text is defined before it is formatted into the profile.

ship-init