The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill loads and processes data from various external files to inform its dashboard design, creating a potential vector for instructions embedded in data to influence the agent.
Ingestion points: The process involves reading multiple local files including ~/.claude-marketing/brands/{slug}/profile.json, ~/.claude-marketing/brands/{slug}/guidelines/_manifest.json, and files within ~/.claude-marketing/sops/.
Boundary markers: The skill lacks explicit instructions or delimiters to isolate data found in these files from the agent's core instructions.
Capability inventory: The skill is primarily a documentation and design tool; it does not exhibit capabilities for network exfiltration, arbitrary command execution, or file writing in its provided instructions.
Sanitization: There is no evidence of sanitization or schema validation performed on the ingested JSON or markdown content before it is processed by the agent.

executive-dashboard