quality
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from the repository workspace under review and possesses a capability tier that allows local command execution.
- Ingestion points: The skill reads changed files in the workspace and analyzes commit data via
git diff --statandgit log --onelineinsideSKILL.md. - Boundary markers: Absent. There are no delimiters or explicit instructions to prevent the model from following guidelines or instructions embedded within the reviewed code files.
- Capability inventory: The skill runs local commands including
go test ./...and executes a workspace shell scriptsh scripts/lint-arch.shinsideSKILL.md. - Sanitization: Absent. No validation or escaping is applied to the workspace files before invoking execution tools.
- [COMMAND_EXECUTION]: The skill invokes several shell commands (
git,go vet,go test,noodle) to perform its quality checks and emit events. Executing tests (go test ./...) or local repository scripts (scripts/lint-arch.sh) on untrusted or unverified code can lead to arbitrary code execution if malicious test hooks or scripts are present in the repository.
Audit Metadata