The Agent Skills Directory

[SAFE]: The skill provides instructional content for Rust CLI development using standard, well-known libraries such as clap, anyhow, thiserror, and assert_cmd.
[SAFE]: Explicit safety directives are included in multiple files (SKILL.md, gemini.md, gpt.md) to prevent the hardcoding of secrets, the disabling of TLS verification for convenience, and the request for sensitive credentials like private keys or tokens from users.
[SAFE]: No malicious obfuscation, unauthorized network operations, or persistence mechanisms were detected. The commands provided (cargo add, cargo test, cargo build) are standard Rust development tools.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to build tools that process user-supplied files. While this creates a theoretical surface for indirect prompt injection if the tool's output is subsequently fed back into an LLM, the skill itself provides boundary-reinforcing instructions like using structured JSON output and separate stdout/stderr streams which help mitigate such risks.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns were found. All external references are documentation or community discussion links used for research purposes.

rust-cli