The Agent Skills Directory

[SAFE]: The skill serves as an extensive documentation library for developers building on Cloudflare. It includes architectural decision trees, API references, and implementation patterns for over 30 Cloudflare services.
[EXTERNAL_DOWNLOADS]: The documentation references standard installation procedures for official Cloudflare tools and libraries (e.g., wrangler, miniflare, @cloudflare/workers-types, @cloudflare/puppeteer) from trusted package registries like NPM and GitHub organizations. These are well-known services and do not contribute to a negative verdict.
[DATA_EXPOSURE_AND_EXFILTRATION]: Code snippets consistently demonstrate best practices for secret management by using environment variables (process.env), Cloudflare Secrets (wrangler secret put), and placeholders (e.g., YOUR_API_TOKEN). No hardcoded credentials or sensitive data leakage patterns were detected.
[COMMAND_EXECUTION]: While the skill documents services like Cloudflare Sandbox and Containers which are designed to execute code, the examples provided are for benign demonstration and administrative management of these services.
[INDIRECT_PROMPT_INJECTION]: The skill provides patterns for building AI agents via the Cloudflare Agents SDK. It accurately describes the attack surface involved when agents ingest untrusted data (such as emails or web content). The provided examples mitigate these risks by demonstrating the use of parameterized SQL queries and structured state management.

cloudflare