wrangler
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches documentation and configuration schemas from official Cloudflare domains to ensure command accuracy.
- Evidence: Links to developers.cloudflare.com are used for retrieving command references and configuration schemas.
- [COMMAND_EXECUTION]: Executes project-local Wrangler CLI commands to manage deployments, resources, and generate TypeScript types.
- Evidence: Instructions include running
wrangler --help,wrangler types, andwrangler deployvia the project's local package manager. - [INDIRECT_PROMPT_INJECTION]: The skill processes local project files which could theoretically contain instructions intended to influence the agent.
- Ingestion points: Reads project configuration files such as
wrangler.jsonc,package.json, andcloudflare.config.ts. - Boundary markers: None explicitly defined for file ingestion, though the skill relies on standard project structures.
- Capability inventory: Shell execution (
wrangler), file system modification (editing config), and network access (fetching documentation). - Sanitization: Includes explicit instructions to prevent secret leakage in logs, command arguments, and chat history.
Audit Metadata