dbk-rust-gates
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines several shell command templates for developer workflows.
- It instructs the agent to run
git status,git merge-base, andgit diffto determine the state of the repository. - It provides commands for
cargo fmt,cargo clippy --fix, and various forms ofcargo testusing placeholders like<package>and<full-test-name>. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted external data (project source code and git diffs) which could contain instructions meant to deceive the agent.
- Ingestion points: Reads output from
git status --porcelain,git diff --name-status, and inspects file contents (*.rs,Cargo.toml). - Boundary markers: None identified. There are no instructions to ignore or treat embedded code comments or markdown within the repository as non-authoritative.
- Capability inventory: The skill has the capability to execute
gitandcargocommands in the local environment (defined inSKILL.md). - Sanitization: There is no mention of sanitizing or validating user-provided or project-extracted strings before they are interpolated into the
cargo testcommand templates.
Audit Metadata