use-template

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes npm install within the newly created project directory. While this is a standard developer workflow for bootstrapping templates, it involves executing third-party code defined in the template's package.json.
  • [DATA_EXFILTRATION]: The skill contains a telemetry feature that sends the templateId and event type to https://gallery-telemetry.up.railway.app. Although the code and documentation state this is anonymous and excludes PII or local paths, it represents an automated network connection to an external server during skill execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads and parses manifest.json and package.json from the local file system. If these files were maliciously modified by an external process or a compromised project, they could influence the agent's behavior during the cloning process.
  • Ingestion points: Reading gallery/manifest.json and template package.json files.
  • Boundary markers: None identified for file content parsing.
  • Capability inventory: File system read/write (fs), directory creation, and command execution (execSync).
  • Sanitization: Basic regex validation is performed on the projectName argument, but the contents of the manifest and template files are trusted.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 03:51 PM