The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external user data via placeholders such as [description], [customer], and [problem] to generate brand narratives. It lacks explicit boundary markers or instructions to prevent the agent from following commands embedded within that untrusted data. While no sensitive capabilities are currently exposed, this pattern represents a vulnerability surface for indirect prompt injection.
[METADATA_POISONING]: There is a significant internal inconsistency between the marketing focus of the StoryBrand framework and the 'What Claude Does' and 'Skill Boundaries' sections, which both describe video production and storyboarding tasks. This misleading metadata indicates a likely copy-paste error that obscures the skill's actual operational limits.
[EXTERNAL_DOWNLOADS]: The skill metadata specifies a dependency on the '@clawfu/mcp-skills' Node.js package, indicating that the skill relies on external code execution through a Model Context Protocol server.

storytelling-storybrand