The Agent Skills Directory

[SAFE]: The skill serves as a reference guide for Cloudflare Workers and Durable Objects. All code snippets are instructional and utilize standard, documented platform features.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or secrets were found. The skill actively promotes security best practices by instructing users to store sensitive information using the wrangler secret put command rather than in plaintext variables.
[DATA_EXFILTRATION]: No unauthorized network operations or data exfiltration patterns were identified. The network calls within the implementation examples are directed to legitimate Cloudflare services (AI Inference, Vectorize, KV).
[COMMAND_EXECUTION]: No dangerous system commands or privilege escalation attempts were found. The documentation references the wrangler CLI, which is the standard development tool for the platform.
[INDIRECT_PROMPT_INJECTION]: While the skill's implementation example demonstrates a pattern for processing user input through an LLM, this is documented within the context of building AI-integrated applications. The skill does not contain instructions that would compromise the agent's own behavior or safety guidelines.

cloudflare-workers