workers-dev-experience

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • REMOTE_CODE_EXECUTION (HIGH): The skill uses a piped shell script for installation (curl -fsSL https://bun.sh/install | bash). This is a high-risk pattern that executes unverified remote code directly in the shell. Although Bun is a reputable tool, its source domain is not listed in the trusted whitelist, maintaining its high severity.
  • COMMAND_EXECUTION (MEDIUM): The skill relies on executing various CLI tools (bunx, wrangler, tsc) which have broad permissions to modify the local environment and interact with the network.
  • EXTERNAL_DOWNLOADS (LOW): Multiple dependencies are downloaded from external registries (NPM via Bun). While these are standard for Cloudflare development, they represent an external dependency chain.
  • PROMPT_INJECTION (LOW): Debugging tools in references/debugging-tools.md log untrusted data from incoming HTTP requests (request.headers, request.cf) without sanitization, establishing a surface for indirect prompt injection or log poisoning. Evidence: 1. Ingestion points: request.headers, request.cf in references/debugging-tools.md. 2. Boundary markers: Absent. 3. Capability inventory: fetch (network), console.log (stdout). 4. Sanitization: Absent.
Recommendations
  • HIGH: Downloads and executes remote code from: https://bun.sh/install - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:05 PM