dev-rust
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content, including source code, manifests, and guidance files, which serves as an ingestion surface. It possesses the capability to perform filesystem edits and execute toolchain commands. Ingestion points: Repository source code, dependency manifests, and project-specific guidance files. Boundary markers: Instructions to read host-provided rules and specific repository guidance files (CLAUDE.md, AGENTS.md). Capability inventory: Source code modification, manifest updates, and execution of repository tools (cargo, git). Sanitization: Reliance on repository-native validation gates, formatters, and linters.
- [COMMAND_EXECUTION]: Executes standard development and toolchain commands such as git and cargo to validate implementations and ensure repository alignment.
- [SAFE]: Includes explicit instructions to exclude secrets and credentials from grounding records and adheres to vendor-owned infrastructure for external resources.
Audit Metadata