The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions were found that attempt to override agent constraints, bypass safety filters, or extract system prompts.
[Data Exposure & Exfiltration] (SAFE): No sensitive file paths (e.g., SSH keys, AWS credentials) are accessed. The skill correctly recommends using environment variables for sensitive configuration like 'DATABASE_URL'.
[Obfuscation] (SAFE): No evidence of Base64 encoding, zero-width characters, homoglyphs, or other obfuscation techniques designed to hide malicious intent.
[Unverifiable Dependencies] (SAFE): The skill references standard, reputable crates from the Rust ecosystem (axum, tokio, sqlx, serde, thiserror, clap). No untrusted remote scripts or installers are utilized.
[Indirect Prompt Injection] (LOW): The skill defines an attack surface by allowing the agent to fetch external library documentation.
Ingestion points: Uses 'mcp__context7__get-library-docs' to pull content from external GitHub repositories.
Boundary markers: No specific delimiters or 'ignore' instructions are defined for the ingested documentation content.
Capability inventory: The skill is restricted to read-only tools ('Read', 'Grep', 'Glob') and documentation helpers. It lacks capabilities for file writing, network exfiltration, or arbitrary command execution.
Sanitization: No explicit sanitization of external documentation is described.
[Privilege Escalation] (SAFE): No use of 'sudo', 'chmod 777', or other commands that attempt to elevate system privileges.

moai-lang-rust