The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access commands were detected in the skill files.
[CREDENTIALS_UNSAFE]: The skill demonstrates secure development practices by explicitly advising against storing secrets in configuration files and recommending the use of encrypted secret management tools.
[INDIRECT_PROMPT_INJECTION]: The skill provides templates for processing external web requests. While this creates an ingestion point for untrusted data, the skill includes standard routing and error handling logic.
Ingestion points: src/index.ts processes Request objects containing headers, URLs, and bodies.
Boundary markers: None present in the code snippets.
Capability inventory: Includes network requests via fetch and database operations via KV and Durable Objects.
Sanitization: Demonstrates basic JSON parsing and response formatting but does not include explicit security-focused sanitization of user input.

cloudflare-worker-dev