The Agent Skills Directory

[PROMPT_INJECTION]: The skill implements a self-improvement mechanism through the references/learnings.md file.
Ingestion points: The agent is instructed to read references/learnings.md at the start of every invocation to gain context from previous sessions.
Boundary markers: No delimiters or protective instructions are present to prevent the agent from obeying instructions that might be maliciously injected into the learnings file.
Capability inventory: The skill provides guidance on executing shell commands (npx) and interacting with an external search API.
Sanitization: No validation or sanitization is performed on the content appended to or read from the learnings file.
[COMMAND_EXECUTION]: The SKILL.md file contains instructions for the user to run npx skills add sales-skills/sales to install additional tools. These commands are associated with the vendor 'sales-skills' and align with the skill's purpose for managing sales-related directory launches.

sales-theresanaiforthat