The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for project verification, including cargo test, cargo clippy, cargo miri test, and make check. These are standard tools in the Rust and C development lifecycles.
[EXTERNAL_DOWNLOADS]: The instructions reference well-known libraries and tools from the Rust ecosystem, such as bindgen, cbindgen, the cc crate, and signal-hook. These dependencies are standard for C-to-Rust interoperability.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external C source code, headers, and build files. This ingestion creates a surface where malicious instructions could be embedded in the code or comments of the project being ported. The risk is relevant because the agent is also instructed to perform command-line build and test operations on this input. 1. Ingestion points: C source files and headers processed for porting (SKILL.md, MAPPING.md). 2. Boundary markers: None explicitly defined to separate user-provided code from agent instructions. 3. Capability inventory: Shell execution for build tools and test runners (cargo, make) described in SKILL.md. 4. Sanitization: No specific sanitization or validation of the input source code is defined before processing or execution.

port-from-c