graphviz

Pass

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: LOW
Full Analysis
  • Prompt Injection (SAFE): The content is strictly limited to Graphviz syntax rules and examples. There are no instructions aimed at overriding system prompts, bypassing safety filters, or extracting the system prompt.
  • Data Exposure & Exfiltration (SAFE): No file access operations, sensitive path references, or network requests (curl, wget) are present. The only URL provided is a documentation link in the metadata.
  • Obfuscation (SAFE): Analysis confirms no Base64 encoding, zero-width characters, homoglyphs, or URL-encoded hidden commands.
  • Indirect Prompt Injection (INFO): The skill defines a surface for processing untrusted data to create diagrams. 1. Ingestion points: User-provided diagram descriptions. 2. Boundary markers: Explicit instructions to use dot code fences. 3. Capability inventory: None (no scripts, file writes, or network access). 4. Sanitization: Not applicable as the skill only generates markdown text. Severity is INFO as the skill only displays data with no side-effect capabilities.
  • Unverifiable Dependencies (SAFE): The skill does not include any requirements files (package.json, requirements.txt) or remote script downloads.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 15, 2026, 09:47 PM