The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions were found that attempt to override agent behavior, bypass safety filters, or extract system prompts. All instructional content is strictly focused on technical Rust refactoring guidelines.
[Data Exposure & Exfiltration] (SAFE): The skill contains no hardcoded credentials (API keys, tokens), sensitive file paths (e.g., .ssh, .aws), or unauthorized network operations.
[Obfuscation] (SAFE): No encoded content (Base64), zero-width characters, homoglyphs, or other obfuscation techniques were detected. The content is entirely human-readable Markdown.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): No remote script execution patterns (e.g., curl|bash) or runtime package installations were identified. While the README mentions developer-side build scripts (pnpm), the skill itself does not execute these during agent operation.
[Privilege Escalation] (SAFE): No commands or patterns associated with acquiring elevated privileges (e.g., sudo, chmod) were found.
[Persistence Mechanisms] (SAFE): No attempts to create persistence (e.g., modifying shell profiles or cron jobs) were detected.
[Metadata Poisoning] (SAFE): Metadata in SKILL.md and metadata.json is accurate, consistent with the skill's stated purpose, and free of malicious instructions.
[Indirect Prompt Injection] (SAFE): The skill is a static knowledge base for code analysis. 1. Ingestion points: Ingests user-provided Rust source code for refactoring. 2. Boundary markers: None defined within the skill itself. 3. Capability inventory: No tool execution, network access, or file-writing capabilities. 4. Sanitization: Not applicable as no code is executed.
[Time-Delayed / Conditional Attacks] (SAFE): No logic gating behavior based on date, time, or environment triggers was found.
[Dynamic Execution] (SAFE): The skill does not generate, load, or execute code at runtime. It provides static reference examples.

rust-idioms