The Agent Skills Directory

[PROMPT_INJECTION]: The skill implements a self-improvement mechanism where it reads from and appends to a persistent memory file (references/learnings.md). This creates an indirect prompt injection surface where maliciously crafted user input or ingested data could be stored and influence the agent's behavior in future sessions.
Ingestion points: The skill reads references/learnings.md at the start of each invocation and accepts arbitrary user input describing Rafiki-related issues.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt templates when processing these inputs.
Capability inventory: The skill primarily generates text-based guidance and routes users to other internal skills (/sales-cadence, /sales-coaching, etc.). It does not possess high-privilege capabilities such as arbitrary shell execution or direct network access.
Sanitization: No input validation or sanitization of the content written to the learnings file is identified.
[EXTERNAL_DOWNLOADS]: The documentation includes instructions for the user to install a related tool using npx skills add sales-skills/sales. This command references a package from the same vendor namespace ('sales-skills'), representing a standard installation procedure for the vendor's ecosystem.

sales-rafiki