cloudflare-pages
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents several shell commands for project setup, CLI authentication, and deployment utilizing the
wranglerCLI and standard tools likecurlandgit. - [EXTERNAL_DOWNLOADS]: The documentation includes instructions to download and install standard developer dependencies and CLI tools from the NPM registry (e.g.,
wrangler,hono,prisma,agents). These originate from well-known and reputable sources. - [INDIRECT_PROMPT_INJECTION]: The skill provides templates for Pages Functions that ingest external data, representing a vulnerability surface if developers do not implement proper sanitization.
- Ingestion points: Serverless function scripts in
SKILL.mdandreferences/framework_guides.mdingest untrusted data from HTTP requests usingrequest.json(). - Boundary markers: The code snippets provide functional templates without explicit security delimiters; however, the skill includes a dedicated 'Security Recommendations' section.
- Capability inventory: Functions demonstrate access to database operations (D1), object storage (R2), and service bindings for external network requests.
- Sanitization: The skill explicitly documents the need to 'Validate and sanitize user input' as a best practice for security.
Audit Metadata