The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external and local sources, establishing an ingestion surface for potential indirect prompt injection attacks.
Ingestion points: Untrusted data enters the agent context through user-provided $ARGUMENTS and local files including founder/facts.md, founder/product-brief.md, founder/persona-gen.md, founder/mvp-scope.md, and founder/pricing-strategy.md (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt template.
Capability inventory: The skill uses WebSearch and WebFetch for network operations and has scoped Read/Edit permissions for the founder/ directory.
Sanitization: There is no evidence of sanitization or validation of the ingested data.
[EXTERNAL_DOWNLOADS]: The skill utilizes WebSearch and WebFetch tools to retrieve marketing benchmarks and performance metrics from the internet. This is a functional requirement for providing data-backed email strategies and does not involve the execution of remote code or the download of unverifiable software packages.

email-sequence