sandbox-sdk
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references official Cloudflare documentation and their public GitHub repository for SDK examples. These sources are well-known and trusted technology services.
- [COMMAND_EXECUTION]: The agent is instructed to use
rg(ripgrep) to search for configuration files and manifest declarations within the project repository. It also prescribes running the project's own validation commands (e.g., typecheck, tests) as part of a verification workflow. These commands are standard for development tasks. - [INDIRECT_PROMPT_INJECTION]: The skill defines a process for ingesting and editing local project files (e.g.,
wrangler.toml,Dockerfile). While this involves processing external data, the instructions include explicit verification steps and version alignment checks to ensure integrity and compliance with official standards.
Audit Metadata