The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and analyze the content of files located in the .changeset directory. This creates a potential surface for indirect prompt injection, where a malicious file in the repository could attempt to override the agent's instructions during the auditing process. * Ingestion points: Content of files in the .changeset directory (referenced in SKILL.md). * Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present when processing the file content. * Capability inventory: The skill utilizes the agent's ability to read local files and perform git commits. * Sanitization: No sanitization or validation of the changeset file content is performed before the agent evaluates it against the provided criteria.

pre-release