The Agent Skills Directory

[Prompt Injection] (SAFE): No attempts to override agent behavior, bypass safety constraints, or extract system prompts were identified in the instructions or metadata.\n- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded credentials, or unauthorized network operations are present. The examples use standard placeholders for domains and local port bindings.\n- [Obfuscation] (SAFE): No encoded strings, base64, zero-width characters, or homoglyphs were found. The content is entirely human-readable.\n- [Unverifiable Dependencies] (SAFE): The skill correctly references the 'salvo' crate with a specific version. No suspicious remote scripts or piped bash installations are used.\n- [Indirect Prompt Injection] (SAFE): This skill provides static documentation and code templates. It does not ingest untrusted runtime data or create a surface for external instructions to influence agent logic.

salvo-cors