The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read project-specific context from a local file, which acts as an ingestion point for untrusted instructions if the repository environment is compromised.
Ingestion points: The file .claude/product-marketing-context.md is read to tailor marketing recommendations.
Boundary markers: The skill does not provide delimiters or instructions to the agent to ignore potentially malicious content within the context file.
Capability inventory: As the skill contains no restrictions on tool usage, the agent's full set of system and network capabilities is available to potential injections.
Sanitization: There are no verification or sanitization steps performed on the ingested content.
[NO_CODE]: The skill consists entirely of markdown instructions and does not include any executable scripts, binaries, or automated configuration changes.

marketing-psychology