jinja2-prompts

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious behavior detected. The skill focuses on template definition and providing security best practices for prompt engineering.
  • [PROMPT_INJECTION]: The skill establishes a surface for indirect prompt injection by defining placeholders for dynamic content injection. 1. Ingestion points: Variable interpolation within templates like analysis.md.j2 using the {{ variable }} syntax. 2. Boundary markers: Instructions specify that templates must be author-written and never user-supplied. 3. Capability inventory: No executable code or scripts are included in this skill; it defines template logic only. 4. Sanitization: Documentation mandates the use of a SandboxedEnvironment and an explicit variable allowlist to prevent unauthorized system access or prompt manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 06:17 PM
Security Audit — agent-trust-hub — jinja2-prompts