Skills
skills/aia-11-hn-mib/mib-mockinterviewaibot/devops/Gen Agent Trust Hub

devops

Fail

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Remote Code Execution (LOW): The SKILL.md file contains the command curl https://sdk.cloud.google.com | bash. This pattern of piping a remote script directly into a shell is usually a high-risk vector. However, because the source is google.com (a Trusted External Source), the severity is downgraded to LOW per the [TRUST-SCOPE-RULE].
  • Indirect Prompt Injection (LOW): The skill provides patterns for 'AI-Powered Web Scrapers' in references/browser-rendering.md and references/cloudflare-workers-advanced.md.
  • Ingestion points: The skill uses page.content() to ingest untrusted data from external websites.
  • Boundary markers: There are no boundary markers or sanitization logic shown to prevent instructions inside scraped HTML from influencing the AI models (llama-3-8b-instruct).
  • Capability inventory: The environment has access to Cloudflare AI bindings and network requests via Puppeteer.
  • Sanitization: No escaping or validation of external content is present before it is passed to the AI model.
  • External Downloads (LOW): The skill references various installers and packages from trusted domains including google.com, npmjs.com, and cloudflare.com. These are acceptable for the skill's intended DevOps purpose.
Recommendations
  • HIGH: Downloads and executes remote code from: https://sdk.cloud.google.com - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 20, 2026, 01:09 AM