wrangler
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes local project files which could contain untrusted data or malicious instructions. * Ingestion points: Interaction with local files occurs through tools like read_file, search_code, and edit_file on files such as wrangler.jsonc and application code. * Boundary markers: The instructions lack requirements for delimiters or specific warnings to prevent the agent from interpreting file content as instructions. * Capability inventory: The skill utilizes execute_command, wrangler_deploy, and write_file to perform its tasks. * Sanitization: Beyond a mention of schema validation for configuration, there are no instructions for sanitizing or escaping content retrieved from files.
- [COMMAND_EXECUTION]: The skill performs shell operations via execute_command to interact with the Wrangler CLI. This is a core function of the skill but provides an execution path for potential injections.
- [EXTERNAL_DOWNLOADS]: The skill references documentation and configuration schemas from official Cloudflare domains.
Audit Metadata