The Agent Skills Directory

[SAFE]: The skill operates within its documented scope, providing platform-specific advice and configuration guidance without any malicious patterns.
[PROMPT_INJECTION]: The skill utilizes a 'learning' feature to store new insights in references/learnings.md. This represents a surface for indirect prompt injection where untrusted user input could influence future agent behavior, though it is categorized as a low-risk functional feature.
Ingestion points: User prompts and the references/learnings.md reference file.
Boundary markers: No specific delimiters or boundary instructions are implemented for reading or writing to the learnings file.
Capability inventory: Local file-append operations and internal routing to other sales-related skills.
Sanitization: No explicit sanitization or content validation is performed on data written to the reference files.
[EXTERNAL_DOWNLOADS]: The skill documentation includes an npx command for installing additional platform tools. This points to the official repository of the vendor ('sales-skills') and follows standard administrative practices for the ecosystem.

sales-podia