The Agent Skills Directory

[NO_CODE]: The skill consists exclusively of markdown instructions and message templates. No executable scripts, binaries, or code files are included, significantly reducing the potential attack surface.
[SAFE]: No malicious patterns, such as hardcoded credentials or unauthorized network exfiltration, were detected. The skill's reference to company-profile.md is a legitimate use of local data to provide necessary context for generating personalized content.
[PROMPT_INJECTION]: The skill processes data gathered from external prospect research, which establishes a surface for indirect prompt injection. However, the analysis determines the risk is negligible because the skill lacks executable capabilities (no subprocess calls or sensitive file-write access), meaning there are no sinks for a potential injection to exploit. The output remains restricted to message drafts for user review.

outreach-writer