visualize-repo

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads and executes the @agent-native/core package from the NPM registry using npx.
  • [COMMAND_EXECUTION]: The skill executes shell commands to initialize, serve, and validate a visual documentation workspace.
  • [DATA_EXFILTRATION]: The verify command transmits local MDX documentation to a remote service for validation. The documentation notes this egress and provides options to use a local validation endpoint.
  • [PROMPT_INJECTION]: The skill reads and processes local repository files, MDX documents, and feedback comments, which creates a surface for indirect prompt injection. * Ingestion points: Repository source code, agent-native.json, and .agent-native/visual-docs/ files. * Boundary markers: No specific delimiters or safety instructions are defined for the agent when reading these files. * Capability inventory: Includes shell command execution and local file system modifications. * Sanitization: The skill does not explicitly describe sanitizing or filtering ingested file content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:51 AM
Security Audit — agent-trust-hub — visualize-repo