The Agent Skills Directory

[SAFE]: The skill consists of instructional prompts to guide an AI agent in a solutions architect role. The instructions focus on project diagnosis, user interviews, and the creation of standard markdown documentation (tareas.md, plan_implementacion.md).
[PROMPT_INJECTION]: The skill features a functional surface for indirect prompt injection as it is instructed to scan existing code in the workspace (Scenario B in SKILL.md). This is a necessary capability for its stated purpose of project evolution.
Ingestion points: Scans workspace files such as src and package.json as defined in SKILL.md.
Boundary markers: No specific boundary delimiters or safety instructions are provided to mitigate risks from malicious content within scanned files.
Capability inventory: The skill materializes its diagnosis by writing markdown files to the local file system.
Sanitization: No explicit sanitization or input validation is implemented for the ingested project data.
[SAFE]: No obfuscation, data exfiltration, or unauthorized command execution patterns were found in the skill content or its metadata.

seo-audit-skill