The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a significant surface area for indirect prompt injection due to its core function of reading and analyzing external project data.
Ingestion points: The agent reads workspace files, specifically Cargo.toml and .rs source files, using the Read, Glob, and Grep tools.
Boundary markers: The instructions do not define delimiters or specify that the agent should ignore instructions embedded within the files it reviews.
Capability inventory: The agent has high-privilege capabilities including the Bash tool for command execution and Write/Edit tools for modifying the file system.
Sanitization: There is no mention of sanitizing or validating ingested code before it is processed by the model.
[COMMAND_EXECUTION]: The skill is configured to use the Bash tool. While this is necessary for standard Rust development tasks such as running cargo test, miri, or cargo-fuzz, it provides a powerful execution environment that could be abused if the agent is compromised via indirect injection from a malicious repository.

rust-engineer