wrangler
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill installs the 'wrangler' package, which is the official command-line interface provided by Cloudflare for its Workers platform. Cloudflare is a well-known and trusted technology company.
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external data sources and local project files that could contain embedded instructions. This is a common surface for development tools.
- Ingestion points: The skill reads local project directories for deployment (
wrangler pages deploy), SQL files for database execution (wrangler d1 execute --file), and remote data from Key-Value storage (KV) and object storage (R2). - Boundary markers: No specific delimiters or instructions are provided to isolate untrusted data from the agent's execution context.
- Capability inventory: The skill allows for local file system access, network communication with Cloudflare APIs, and management of sensitive environment secrets.
- Sanitization: There are no built-in mechanisms described to sanitize or validate the content of files or database records before they are processed by the agent.
Audit Metadata