The Agent Skills Directory

[NO_CODE]: The analyzed skill is composed strictly of Markdown documentation and templates. It contains no executable Python, JavaScript, or shell scripts.
Files analyzed include: SKILL.md, ARCHITECTURE_REVIEW_TEMPLATE.md, COST_OPTIMIZATION_REPORT_TEMPLATE.md, and SCALABILITY_ANALYSIS_TEMPLATE.md.
[SAFE]: No malicious patterns or threat vectors were detected across any of the analyzed files.
Prompt Injection: The instructions do not contain overrides, jailbreaks, or attempts to bypass system constraints. It actually emphasizes guardrails like cost ceilings and loop limits.
Data Exfiltration: There are no network operations (curl, wget, fetch) or access to sensitive local file paths (e.g., .ssh, .aws).
Obfuscation: No Base64 encoding, zero-width characters, or homoglyph attacks were identified in the text or metadata.
[SAFE]: The skill demonstrates positive security practices by providing explicit instructions for PII masking and verifying that RAG sources do not expose unauthorized documents.

performant-ai