The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from email-profile.md to customize strategy templates. This creates a potential surface for indirect prompt injection if the profile file contains malicious instructions designed to hijack the agent's behavior during strategy generation.
Ingestion point: email-profile.md is read during the business type detection workflow defined in SKILL.md.
Boundary markers: No delimiters or specific boundary instructions are used when reading and processing the profile content.
Capability inventory: The skill utilizes Read, Write, Grep, and Glob tools to manage strategy assets and generate the final implementation roadmap.
Sanitization: The skill does not perform validation or sanitization of the input file content before using it to select templates or populate the strategy document.

email-plan