The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch platform-specific documentation from the vendor's GitHub repository (github.com/sales-skills/) and other potentially configured third-party repositories. This is used to ground recommendations in concrete, up-to-date documentation.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes instructions from external Markdown files (SKILL.md from other repositories) and local learning files without explicit boundary markers or sanitization of the fetched content. The mandatory evidence chain is as follows:
Ingestion points: Reads references/learnings.md, references/platforms.md, and remote SKILL.md files via WebFetch.
Boundary markers: Absent.
Capability inventory: WebFetch and Read operations followed by text generation and skill invocation recommendations.
Sanitization: Absent.
[NO_CODE]: The skill consists entirely of instructional Markdown and reference files; no executable scripts, binaries, or configuration files are included in the distribution.

sales-deliverability