The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and strictly follow instructions from a local brand-context.md file, which is explicitly allowed to override the agent's default behavior and constraints.
Ingestion points: The skill searches for brand-context.md in the project root, ./.claude/, or ./docs/ directories.
Boundary markers: The instructions tell the agent to "read it in full and treat it as binding" and that its rules "override an agent's own defaults."
Capability inventory: The skill is primarily used for routing tasks and generating marketing deliverables; no dangerous tools or system-level capabilities were found in the provided files.
Sanitization: No sanitization or validation of the ingested file content is performed; the agent is instructed to trust the file as the primary source of truth and rules.

events-ops