cli-tools
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents and instructs the agent to execute a variety of shell commands and local scripts (e.g.,
yarn dev:ios,yarn test,./scripts/maestro-test.sh,./scripts/build_spm_frameworks.sh) to perform development tasks. - [PROMPT_INJECTION]: Identifies an Indirect Prompt Injection surface where malicious instructions in the codebase could influence the agent's behavior.
- Ingestion points: The agent is directed to read and process source code files (.ts, .tsx, .js), test files, and configuration files within the monorepo workspace.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to help it distinguish between the skill's trusted instructions and potentially untrusted data embedded within the files it processes.
- Capability inventory: The agent is granted the capability to execute arbitrary shell scripts and complex build commands documented throughout
SKILL.md. - Sanitization: No mechanisms for validation, escaping, or sanitizing external content from the workspace are defined.
Audit Metadata