spezi-platform-selection

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local shell script (scripts/clone-template.sh) to perform repository cloning. This script uses the standard git clone command to retrieve template code.
  • [EXTERNAL_DOWNLOADS]: The skill downloads project templates from the official Stanford Spezi GitHub repositories (github.com/StanfordSpezi). These are legitimate vendor resources used for application scaffolding.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by ingesting and moving user-provided planning documents (docs/planning/ and docs/implementation-plan.md) into the new project directory.
  • Ingestion points: Reads files from the docs/planning/ directory and the docs/implementation-plan.md file in the current working directory.
  • Boundary markers: The instructions explicitly state to "Confirm with the user before moving" which provides a manual checkpoint.
  • Capability inventory: The skill uses the mv command for file system operations and executes git clone via a shell script to fetch remote data.
  • Sanitization: No specific sanitization or validation of the contents within the planning documents is performed prior to the move operation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:54 PM