The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized network operations were detected. The skill is constrained to using standard file system tools (Read, Glob, Write, Grep) within the '.chalk/docs/engineering/' directory for its intended purpose of documentation management.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted content from existing project documents and user arguments to generate new files. * Ingestion points: Reads existing ADRs and engineering docs from '.chalk/docs/engineering/' and accepts '$ARGUMENTS' from the user. * Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore instructions embedded within the source documents. * Capability inventory: The skill uses the 'Write' tool to create markdown files on the local filesystem. * Sanitization: No sanitization or validation of the ingested documentation or user input is performed before the content is drafted into the new ADR.

create-adr